Introduction to Parallel Computing 2nd Edition

Ananth Grama download pdf

https://ebookultra.com/download/introduction-to-parallel-computing-2nd-
edition-ananth-grama/

Visit ebookultra.com today to download the complete set of

ebook or textbook!
We have selected some products that you may be interested in
Click the link to download now or visit ebookultra.com
for more options!.

Introduction to Parallel Computing 2nd Edition Ananth

Grama

https://ebookultra.com/download/introduction-to-parallel-
computing-2nd-edition-ananth-grama-2/

An introduction to parallel and vector scientific

computing 1st Edition Ronald W. Shonkwiler

https://ebookultra.com/download/an-introduction-to-parallel-and-
vector-scientific-computing-1st-edition-ronald-w-shonkwiler/

Elements of Parallel Computing 1st Edition Eric Aubanel

https://ebookultra.com/download/elements-of-parallel-computing-1st-
edition-eric-aubanel/

Introduction to grid computing with Globus 2nd ed Edition

Luis Ferreira

https://ebookultra.com/download/introduction-to-grid-computing-with-
globus-2nd-ed-edition-luis-ferreira/
Parallel Computing in Quantum Chemistry 1st Edition Curtis
L. Janssen

https://ebookultra.com/download/parallel-computing-in-quantum-
chemistry-1st-edition-curtis-l-janssen/

High Performance Parallel Database Processing and Grid

Databases Wiley Series on Parallel and Distributed
Computing 1st Edition David Taniar
https://ebookultra.com/download/high-performance-parallel-database-
processing-and-grid-databases-wiley-series-on-parallel-and-
distributed-computing-1st-edition-david-taniar/

An Introduction to Quantum Computing 1st Edition Phillip

Kaye

https://ebookultra.com/download/an-introduction-to-quantum-
computing-1st-edition-phillip-kaye/

The dRuby Book Distributed and Parallel Computing with

Ruby 1st Edition Masatoshi Seki

https://ebookultra.com/download/the-druby-book-distributed-and-
parallel-computing-with-ruby-1st-edition-masatoshi-seki/

Mobile Intelligence Wiley Series on Parallel and

Distributed Computing 1st Edition Laurence T. Yang

https://ebookultra.com/download/mobile-intelligence-wiley-series-on-
parallel-and-distributed-computing-1st-edition-laurence-t-yang/
Introduction to Parallel Computing 2nd Edition Ananth
Grama Digital Instant Download
Author(s): Ananth Grama
ISBN(s): 9780849344961, 0849344964
Edition: 2
File Details: PDF, 12.32 MB
Year: 2003
Language: english
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks .

[ Team LiB ]

• Table of Contents
Introduction to Parallel Computing, Second Edition

By Ananth Grama, Anshul Gupta, George Karypis, Vipin Kumar

Publisher: Addison Wesley

Pub Date: January 16, 2003

ISBN: 0-201-64865-2

Pages: 856

Increasingly, parallel processing is being seen as the only cost-effective method for the fast solution of computationally large and
data-intensive problems. The emergence of inexpensive parallel computers such as commodity desktop multiprocessors and clusters of
workstations or PCs has made such parallel methods generally applicable, as have software standards for portable parallel
programming. This sets the stage for substantial growth in parallel software.

Data-intensive applications such as transaction processing and information retrieval, data mining and analysis and multimedia services
have provided a new challenge for the modern generation of parallel platforms. Emerging areas such as computational biology and
nanotechnology have implications for algorithms and systems development, while changes in architectures, programming models and
applications have implications for how parallel platforms are made available to users in the form of grid-based services.

This book takes into account these new developments as well as covering the more traditional problems addressed by parallel
computers.Where possible it employs an architecture-independent view of the underlying platforms and designs algorithms for an
abstract model. Message Passing Interface (MPI), POSIX threads and OpenMP have been selected as programming models and the
evolving application mix of parallel computing is reflected in various examples throughout the book.

[ Team LiB ]
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks.

[ Team LiB ]

• Table of Contents
Introduction to Parallel Computing, Second Edition

By Ananth Grama, Anshul Gupta, George Karypis, Vipin Kumar

Publisher: Addison Wesley

Pub Date: January 16, 2003
ISBN: 0-201-64865-2
Pages: 856

Copyright
Pearson Education
Preface
Acknowledgments
Chapter 1. Introduction to Parallel Computing
Section 1.1. Motivating Parallelism
Section 1.2. Scope of Parallel Computing
Section 1.3. Organization and Contents of the Text
Section 1.4. Bibliographic Remarks
Problems
Chapter 2. Parallel Programming Platforms
Section 2.1. Implicit Parallelism: Trends in Microprocessor Architectures*
Section 2.2. Limitations of Memory System Performance*
Section 2.3. Dichotomy of Parallel Computing Platforms
Section 2.4. Physical Organization of Parallel Platforms
Section 2.5. Communication Costs in Parallel Machines
Section 2.6. Routing Mechanisms for Interconnection Networks
Section 2.7. Impact of Process-Processor Mapping and Mapping Techniques
Section 2.8. Bibliographic Remarks
Problems
Chapter 3. Principles of Parallel Algorithm Design
Section 3.1. Preliminaries
Section 3.2. Decomposition Techniques
Section 3.3. Characteristics of Tasks and Interactions
Section 3.4. Mapping Techniques for Load Balancing
Section 3.5. Methods for Containing Interaction Overheads
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks.

Section 3.6. Parallel Algorithm Models

Section 3.7. Bibliographic Remarks
Problems
Chapter 4. Basic Communication Operations
Section 4.1. One-to-All Broadcast and All-to-One Reduction
Section 4.2. All-to-All Broadcast and Reduction
Section 4.3. All-Reduce and Prefix-Sum Operations
Section 4.4. Scatter and Gather
Section 4.5. All-to-All Personalized Communication
Section 4.6. Circular Shift
Section 4.7. Improving the Speed of Some Communication Operations
Section 4.8. Summary
Section 4.9. Bibliographic Remarks
Problems
Chapter 5. Analytical Modeling of Parallel Programs
Section 5.1. Sources of Overhead in Parallel Programs
Section 5.2. Performance Metrics for Parallel Systems
Section 5.3. The Effect of Granularity on Performance
Section 5.4. Scalability of Parallel Systems
Section 5.5. Minimum Execution Time and Minimum Cost-Optimal Execution Time
Section 5.6. Asymptotic Analysis of Parallel Programs
Section 5.7. Other Scalability Metrics
Section 5.8. Bibliographic Remarks
Problems
Chapter 6. Programming Using the Message-Passing Paradigm
Section 6.1. Principles of Message-Passing Programming
Section 6.2. The Building Blocks: Send and Receive Operations
Section 6.3. MPI: the Message Passing Interface
Section 6.4. Topologies and Embedding
Section 6.5. Overlapping Communication with Computation
Section 6.6. Collective Communication and Computation Operations
Section 6.7. Groups and Communicators
Section 6.8. Bibliographic Remarks
Problems
Chapter 7. Programming Shared Address Space Platforms
Section 7.1. Thread Basics
Section 7.2. Why Threads?
Section 7.3. The POSIX Thread API
Section 7.4. Thread Basics: Creation and Termination
Section 7.5. Synchronization Primitives in Pthreads
Section 7.6. Controlling Thread and Synchronization Attributes
Section 7.7. Thread Cancellation
Section 7.8. Composite Synchronization Constructs
Section 7.9. Tips for Designing Asynchronous Programs
Section 7.10. OpenMP: a Standard for Directive Based Parallel Programming
Section 7.11. Bibliographic Remarks
Problems
Chapter 8. Dense Matrix Algorithms
Section 8.1. Matrix-Vector Multiplication
Section 8.2. Matrix-Matrix Multiplication
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks.

Section 8.3. Solving a System of Linear Equations

Section 8.4. Bibliographic Remarks
Problems
Chapter 9. Sorting
Section 9.1. Issues in Sorting on Parallel Computers
Section 9.2. Sorting Networks
Section 9.3. Bubble Sort and its Variants
Section 9.4. Quicksort
Section 9.5. Bucket and Sample Sort
Section 9.6. Other Sorting Algorithms
Section 9.7. Bibliographic Remarks
Problems
Chapter 10. Graph Algorithms
Section 10.1. Definitions and Representation
Section 10.2. Minimum Spanning Tree: Prim's Algorithm
Section 10.3. Single-Source Shortest Paths: Dijkstra's Algorithm
Section 10.4. All-Pairs Shortest Paths
Section 10.5. Transitive Closure
Section 10.6. Connected Components
Section 10.7. Algorithms for Sparse Graphs
Section 10.8. Bibliographic Remarks
Problems
Chapter 11. Search Algorithms for Discrete Optimization Problems
Section 11.1. Definitions and Examples
Section 11.2. Sequential Search Algorithms
Section 11.3. Search Overhead Factor
Section 11.4. Parallel Depth-First Search
Section 11.5. Parallel Best-First Search
Section 11.6. Speedup Anomalies in Parallel Search Algorithms
Section 11.7. Bibliographic Remarks
Problems
Chapter 12. Dynamic Programming
Section 12.1. Overview of Dynamic Programming
Section 12.2. Serial Monadic DP Formulations
Section 12.3. Nonserial Monadic DP Formulations
Section 12.4. Serial Polyadic DP Formulations
Section 12.5. Nonserial Polyadic DP Formulations
Section 12.6. Summary and Discussion
Section 12.7. Bibliographic Remarks
Problems
Chapter 13. Fast Fourier Transform
Section 13.1. The Serial Algorithm
Section 13.2. The Binary-Exchange Algorithm
Section 13.3. The Transpose Algorithm
Section 13.4. Bibliographic Remarks
Problems
Appendix A. Complexity of Functions and Order Analysis
Section A.1. Complexity of Functions
Section A.2. Order Analysis of Functions
Bibliography
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks .

[ Team LiB ]
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks .

[ Team LiB ]
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks

Copyright
Pearson Education Limited

Edinburgh Gate

Harlow

Essex CM20 2JE

England

and Associated Companies throughout the world

Visit us on the World Wide Web at: www.pearsoneduc.com

First published by The Benjamin/Cummings Publishing Company, Inc. 1994

Second edition published 2003

© The Benjamin/Cummings Publishing Company, Inc. 1994

© Pearson Education Limited 2003

The rights of Ananth Grama, Anshul Gupta, George Karypis and Vipin Kumar to be identified as authors of this work have been asserted
by them in accordance with the Copyright, Designs and Patents Act 1988.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any
means, electronic, mechanical, photocopying, recording or otherwise, without either the prior written permission of the publisher or a
licence permitting restricted copying in the United Kingdom issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road,
London W1T 4LP.

The programs in this book have been included for their instructional value. They have been tested with care but are not guaranteed for
any particular purpose. The publisher does not offer any warranties or representations nor does it accept any liabilities with respect to
the programs.

All trademarks used herein are the property of their respective owners. The use of any trademark in this text does not vest in the author
or publisher any trademark ownership rights in such trademarks, nor does the use of such trademarks imply any affiliation with or
endorsement of this book by such owners.

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress

10 9 8 7 6 5 4 3 2 1

07 06 05 04 03

Printed and bound in the United States of America

Dedication
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks .

To Joanna, Rinku, Krista, and Renu

[ Team LiB ]
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks .

[ Team LiB ]

Pearson Education

We work with leading authors to develop the strongest educational materials in computing, bringing cutting-edge thinking and best
learning practice to a global market.

Under a range of well-known imprints, including Addison-Wesley, we craft high-quality print and electronic publications which help
readers to understand and apply their content, whether studying or at work.

To find out more about the complete range of our publishing, please visit us on the World Wide Web at:
www.pearsoneduc.com

[ Team LiB ]
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks .

[ Team LiB ]

Preface
Since the 1994 release of the text "Introduction to Parallel Computing: Design and Analysis of Algorithms" by the same authors, the field
of parallel computing has undergone significant changes. Whereas tightly coupled scalable message-passing platforms were the norm a
decade ago, a significant portion of the current generation of platforms consists of inexpensive clusters of workstations, and
multiprocessor workstations and servers. Programming models for these platforms have also evolved over this time. Whereas most
machines a decade back relied on custom APIs for messaging and loop-based parallelism, current models standardize these APIs
across platforms. Message passing libraries such as PVM and MPI, thread libraries such as POSIX threads, and directive based models
such as OpenMP are widely accepted as standards, and have been ported to a variety of platforms.

With respect to applications, fluid dynamics, structural mechanics, and signal processing formed dominant applications a decade back.
These applications continue to challenge the current generation of parallel platforms. However, a variety of new applications have also
become important. These include data-intensive applications such as transaction processing and information retrieval, data mining and
analysis, and multimedia services. Applications in emerging areas of computational biology and nanotechnology pose tremendous
challenges for algorithms and systems development. Changes in architectures, programming models, and applications are also being
accompanied by changes in how parallel platforms are made available to the users in the form of grid-based services.

This evolution has a profound impact on the process of design, analysis, and implementation of parallel algorithms. Whereas the
emphasis of parallel algorithm design a decade back was on precise mapping of tasks to specific topologies such as meshes and
hypercubes, current emphasis is on programmability and portability, both from points of view of algorithm design and implementation. To
this effect, where possible, this book employs an architecture independent view of the underlying platforms and designs algorithms for an
abstract model. With respect to programming models, Message Passing Interface (MPI), POSIX threads, and OpenMP have been
selected. The evolving application mix for parallel computing is also reflected in various examples in the book.

This book forms the basis for a single concentrated course on parallel computing or a two-part sequence. Some suggestions for such a
two-part sequence are:

1. Introduction to Parallel Computing: Chapters 1–6. This course would provide the basics of algorithm design and parallel
programming.

2. Design and Analysis of Parallel Algorithms: Chapters 2 and 3 followed by Chapters8–12. This course would provide an
in-depth coverage of design and analysis of various parallel algorithms.

The material in this book has been tested in Parallel Algorithms and Parallel Computing courses at the University of Minnesota and
Purdue University. These courses are taken primarily by graduate students and senior-level undergraduate students in Computer
Science. In addition, related courses in Scientific Computation, for which this material has also been tested, are taken by graduate
students in science and engineering, who are interested in solving computationally intensive problems.

Most chapters of the book include (i) examples and illustrations; (ii) problems that supplement the text and test students' understanding
of the material; and (iii) bibliographic remarks to aid researchers and students interested in learning more about related and advanced
topics. The comprehensive subject index helps the reader locate terms they might be interested in. The page number on which a term is
defined is highlighted in boldface in the index. Furthermore, the term itself appears in bold italics where it is defined. The sections that
deal with relatively complex material are preceded by a '*'. An instructors' manual containing slides of the figures and solutions to
selected problems is also available from the publisher (http://www.booksites.net/kumar).

As with our previous book, we view this book as a continually evolving resource. We thank all the readers who have kindly shared
critiques, opinions, problems, code, and other information relating to our first book. It is our sincere hope that we can continue this
interaction centered around this new book. We encourage readers to address communication relating to this book to
[email protected]. All relevant reader input will be added to the information archived at the sitehttp://www.cs.umn.edu/~parbook with
due credit to (and permission of) the sender(s). An on-line errata of the book will also be maintained at the site. We believe that in a
highly dynamic field such as ours, a lot is to be gained from a healthy exchange of ideas and material in this manner.

[ Team LiB ]
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks .

[ Team LiB ]

Acknowledgments
We would like to begin by acknowledging our spouses, Joanna, Rinku, Krista, and Renu to whom this book is dedicated. Without their
sacrifices this project would not have been seen completion. We also thank our parents, and family members, Akash, Avi, Chethan,
Eleni, Larry, Mary-Jo, Naina, Petros, Samir, Subhasish, Varun, Vibhav, and Vipasha for their affectionate support and encouragement
throughout this project.

Our respective institutions, Computer Sciences and Computing Research Institute (CRI) at Purdue University, Department of Computer
Science & Engineering, the Army High Performance Computing Research Center (AHPCRC), and the Digital Technology Center (DTC)
at the University of Minnesota, and the IBM T. J. Watson Research Center at Yorktown Heights, provided computing resources and
active and nurturing environments for the completion of this project.

This project evolved from our first book. We would therefore like to acknowledge all of the people who helped us with both editions. Many
people contributed to this project in different ways. We would like to thank Ahmed Sameh for his constant encouragement and support,
and Dan Challou, Michael Heath, Dinesh Mehta, Tom Nurkkala, Paul Saylor, and Shang-Hua Teng for the valuable input they provided
to the various versions of the book. We thank the students of the introduction to parallel computing classes at the University of
Minnesota and Purdue university for identifying and working through the errors in the early drafts of the book. In particular, we
acknowledge the patience and help of Jim Diehl and Rasit Eskicioglu, who worked through several early drafts of the manuscript to
identify numerous errors. Ramesh Agarwal, David Bailey, Rupak Biswas, Jim Bottum, Thomas Downar, Rudolf Eigenmann, Sonia
Fahmy, Greg Frederickson, John Gunnels, Fred Gustavson, Susanne Hambrusch, Bruce Hendrickson, Christoph Hoffmann, Kai Hwang,
Ioannis Ioannidis, Chandrika Kamath, David Keyes, Mehmet Koyuturk, Piyush Mehrotra, Zhiyuan Li, Jens Palsberg, Voicu Popescu, Alex
Pothen, Viktor Prasanna, Sanjay Ranka, Naren Ramakrishnan, Elisha Sacks, Vineet Singh, Sartaj Sahni, Vivek Sarin, Wojciech
Szpankowski, Srikanth Thirumalai, Jan Vitek, and David Yau have been great technical resources. It was a pleasure working with the
cooperative and helpful staff at Pearson Education. In particular, we would like to thank Keith Mansfield and Mary Lince for their
professional handling of the project.

The Army Research Laboratory, ARO, DOE, NASA, and NSF provided parallel computing research support for Ananth Grama, George
Karypis, and Vipin Kumar. In particular, Kamal Abdali, Michael Coyle, Jagdish Chandra, Frederica Darema, Stephen Davis, Wm
Randolph Franklin, Richard Hirsch, Charles Koelbel, Raju Namburu, N. Radhakrishnan, John Van Rosendale, Subhash Saini, and
Xiaodong Zhang have been supportive of our research programs in the area of parallel computing. Andrew Conn, Brenda Dietrich, John
Forrest, David Jensen, and Bill Pulleyblank at IBM supported the work of Anshul Gupta over the years.

[ Team LiB ]
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks .

[ Team LiB ]

Chapter 1. Introduction to Parallel Computing

The past decade has seen tremendous advances in microprocessor technology. Clock rates of processors have increased from about
40 MHz (e.g., a MIPS R3000, circa 1988) to over 2.0 GHz (e.g., a Pentium 4, circa 2002). At the same time, processors are now capable
of executing multiple instructions in the same cycle. The average number of cycles per instruction (CPI) of high end processors has
improved by roughly an order of magnitude over the past 10 years. All this translates to an increase in the peak floating point operation
execution rate (floating point operations per second, or FLOPS) of several orders of magnitude. A variety of other issues have also
become important over the same period. Perhaps the most prominent of these is the ability (or lack thereof) of the memory system to
feed data to the processor at the required rate. Significant innovations in architecture and software have addressed the alleviation of
bottlenecks posed by the datapath and the memory.

The role of concurrency in accelerating computing elements has been recognized for several decades. However, their role in providing
multiplicity of datapaths, increased access to storage elements (both memory and disk), scalable performance, and lower costs is
reflected in the wide variety of applications of parallel computing. Desktop machines, engineering workstations, and compute servers
with two, four, or even eight processors connected together are becoming common platforms for design applications. Large scale
applications in science and engineering rely on larger configurations of parallel computers, often comprising hundreds of processors.
Data intensive platforms such as database or web servers and applications such as transaction processing and data mining often use
clusters of workstations that provide high aggregate disk bandwidth. Applications in graphics and visualization use multiple rendering
pipes and processing elements to compute and render realistic environments with millions of polygons in real time. Applications requiring
high availability rely on parallel and distributed platforms for redundancy. It is therefore extremely important, from the point of view of
cost, performance, and application requirements, to understand the principles, tools, and techniques for programming the wide variety of
parallel platforms currently available.

[ Team LiB ]
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks

[ Team LiB ]

1.1 Motivating Parallelism

Development of parallel software has traditionally been thought of as time and effort intensive. This can be largely attributed to the
inherent complexity of specifying and coordinating concurrent tasks, a lack of portable algorithms, standardized environments, and
software development toolkits. When viewed in the context of the brisk rate of development of microprocessors, one is tempted to
question the need for devoting significant effort towards exploiting parallelism as a means of accelerating applications. After all, if it takes
two years to develop a parallel application, during which time the underlying hardware and/or software platform has become obsolete,
the development effort is clearly wasted. However, there are some unmistakable trends in hardware design, which indicate that
uniprocessor (or implicitly parallel) architectures may not be able to sustain the rate of realizable performance increments in the future.
This is a result of lack of implicit parallelism as well as other bottlenecks such as the datapath and the memory. At the same time,
standardized hardware interfaces have reduced the turnaround time from the development of a microprocessor to a parallel machine
based on the microprocessor. Furthermore, considerable progress has been made in standardization of programming environments to
ensure a longer life-cycle for parallel applications. All of these present compelling arguments in favor of parallel computing platforms.

1.1.1 The Computational Power Argument – from Transistors to FLOPS

In 1965, Gordon Moore made the following simple observation:

"The complexity for minimum component costs has increased at a rate of roughly a factor of two per year.
Certainly over the short term this rate can be expected to continue, if not to increase. Over the longer term, the
rate of increase is a bit more uncertain, although there is no reason to believe it will not remain nearly constant for
at least 10 years. That means by 1975, the number of components per integrated circuit for minimum cost will be
65,000."

His reasoning was based on an empirical log-linear relationship between device complexity and time, observed over three data points.
He used this to justify that by 1975, devices with as many as 65,000 components would become feasible on a single silicon chip
occupying an area of only about one-fourth of a square inch. This projection turned out to be accurate with the fabrication of a 16K CCD
memory with about 65,000 components in 1975. In a subsequent paper in 1975, Moore attributed the log-linear relationship to
exponential behavior of die sizes, finer minimum dimensions, and "circuit and device cleverness". He went on to state that:

"There is no room left to squeeze anything out by being clever. Going forward from here we have to depend on
the two size factors - bigger dies and finer dimensions."

He revised his rate of circuit complexity doubling to 18 months and projected from 1975 onwards at this reduced rate. This curve came to
be known as "Moore's Law". Formally, Moore's Law states that circuit complexity doubles every eighteen months. This empirical
relationship has been amazingly resilient over the years both for microprocessors as well as for DRAMs. By relating component density
and increases in die-size to the computing power of a device, Moore's law has been extrapolated to state that the amount of computing
power available at a given cost doubles approximately every 18 months.

The limits of Moore's law have been the subject of extensive debate in the past few years. Staying clear of this debate, the issue of
translating transistors into useful OPS (operations per second) is the critical one. It is possible to fabricate devices with very large
transistor counts. How we use these transistors to achieve increasing rates of computation is the key architectural challenge. A logical
recourse to this is to rely on parallelism – both implicit and explicit. We will briefly discuss implicit parallelism in Section 2.1 and devote
the rest of this book to exploiting explicit parallelism.
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks .

1.1.2 The Memory/Disk Speed Argument

The overall speed of computation is determined not just by the speed of the processor, but also by the ability of the memory system to
feed data to it. While clock rates of high-end processors have increased at roughly 40% per year over the past decade, DRAM access
times have only improved at the rate of roughly 10% per year over this interval. Coupled with increases in instructions executed per clock
cycle, this gap between processor speed and memory presents a tremendous performance bottleneck. This growing mismatch between
processor speed and DRAM latency is typically bridged by a hierarchy of successively faster memory devices called caches that rely on
locality of data reference to deliver higher memory system performance. In addition to the latency, the net effective bandwidth between
DRAM and the processor poses other problems for sustained computation rates.

The overall performance of the memory system is determined by the fraction of the total memory requests that can be satisfied from the
cache. Memory system performance is addressed in greater detail in Section 2.2. Parallel platforms typically yield better memory system
performance because they provide (i) larger aggregate caches, and (ii) higher aggregate bandwidth to the memory system (both typically
linear in the number of processors). Furthermore, the principles that are at the heart of parallel algorithms, namely locality of data
reference, also lend themselves to cache-friendly serial algorithms. This argument can be extended to disks where parallel platforms can
be used to achieve high aggregate bandwidth to secondary storage. Here, parallel algorithms yield insights into the development of
out-of-core computations. Indeed, some of the fastest growing application areas of parallel computing in data servers (database servers,
web servers) rely not so much on their high aggregate computation rates but rather on the ability to pump data out at a faster rate.

1.1.3 The Data Communication Argument

As the networking infrastructure evolves, the vision of using the Internet as one large heterogeneous parallel/distributed computing
environment has begun to take shape. Many applications lend themselves naturally to such computing paradigms. Some of the most
impressive applications of massively parallel computing have been in the context of wide-area distributed platforms. The SETI (Search
for Extra Terrestrial Intelligence) project utilizes the power of a large number of home computers to analyze electromagnetic signals from
outer space. Other such efforts have attempted to factor extremely large integers and to solve large discrete optimization problems.

In many applications there are constraints on the location of data and/or resources across the Internet. An example of such an
application is mining of large commercial datasets distributed over a relatively low bandwidth network. In such applications, even if the
computing power is available to accomplish the required task without resorting to parallel computing, it is infeasible to collect the data at
a central location. In these cases, the motivation for parallelism comes not just from the need for computing resources but also from the
infeasibility or undesirability of alternate (centralized) approaches.

[ Team LiB ]
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks .

[ Team LiB ]
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks

1.2 Scope of Parallel Computing

Parallel computing has made a tremendous impact on a variety of areas ranging from computational simulations for scientific and
engineering applications to commercial applications in data mining and transaction processing. The cost benefits of parallelism coupled
with the performance requirements of applications present compelling arguments in favor of parallel computing. We present a small
sample of the diverse applications of parallel computing.

1.2.1 Applications in Engineering and Design

Parallel computing has traditionally been employed with great success in the design of airfoils (optimizing lift, drag, stability), internal
combustion engines (optimizing charge distribution, burn), high-speed circuits (layouts for delays and capacitive and inductive effects),
and structures (optimizing structural integrity, design parameters, cost, etc.), among others. More recently, design of
microelectromechanical and nanoelectromechanical systems (MEMS and NEMS) has attracted significant attention. While most
applications in engineering and design pose problems of multiple spatial and temporal scales and coupled physical phenomena, in the
case of MEMS/NEMS design these problems are particularly acute. Here, we often deal with a mix of quantum phenomena, molecular
dynamics, and stochastic and continuum models with physical processes such as conduction, convection, radiation, and structural
mechanics, all in a single system. This presents formidable challenges for geometric modeling, mathematical modeling, and algorithm
development, all in the context of parallel computers.

Other applications in engineering and design focus on optimization of a variety of processes. Parallel computers have been used to
solve a variety of discrete and continuous optimization problems. Algorithms such as Simplex, Interior Point Method for linear
optimization and Branch-and-bound, and Genetic programming for discrete optimization have been efficiently parallelized and are
frequently used.

1.2.2 Scientific Applications

The past few years have seen a revolution in high performance scientific computing applications. The sequencing of the human genome
by the International Human Genome Sequencing Consortium and Celera, Inc. has opened exciting new frontiers in bioinformatics.
Functional and structural characterization of genes and proteins hold the promise of understanding and fundamentally influencing
biological processes. Analyzing biological sequences with a view to developing new drugs and cures for diseases and medical
conditions requires innovative algorithms as well as large-scale computational power. Indeed, some of the newest parallel computing
technologies are targeted specifically towards applications in bioinformatics.

Advances in computational physics and chemistry have focused on understanding processes ranging in scale from quantum phenomena
to macromolecular structures. These have resulted in design of new materials, understanding of chemical pathways, and more efficient
processes. Applications in astrophysics have explored the evolution of galaxies, thermonuclear processes, and the analysis of extremely
large datasets from telescopes. Weather modeling, mineral prospecting, flood prediction, etc., rely heavily on parallel computers and
have very significant impact on day-to-day life.

Bioinformatics and astrophysics also present some of the most challenging problems with respect to analyzing extremely large datasets.
Protein and gene databases (such as PDB, SwissProt, and ENTREZ and NDB) along with Sky Survey datasets (such as the Sloan
Digital Sky Surveys) represent some of the largest scientific datasets. Effectively analyzing these datasets requires tremendous
computational power and holds the key to significant scientific discoveries.

1.2.3 Commercial Applications

This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks .

With the widespread use of the web and associated static and dynamic content, there is increasing emphasis on cost-effective servers
capable of providing scalable performance. Parallel platforms ranging from multiprocessors to linux clusters are frequently used as web
and database servers. For instance, on heavy volume days, large brokerage houses on Wall Street handle hundreds of thousands of
simultaneous user sessions and millions of orders. Platforms such as IBMs SP supercomputers and Sun Ultra HPC servers power these
business-critical sites. While not highly visible, some of the largest supercomputing networks are housed on Wall Street.

The availability of large-scale transaction data has also sparked considerable interest in data mining and analysis for optimizing business
and marketing decisions. The sheer volume and geographically distributed nature of this data require the use of effective parallel
algorithms for such problems as association rule mining, clustering, classification, and time-series analysis.

1.2.4 Applications in Computer Systems

As computer systems become more pervasive and computation spreads over the network, parallel processing issues become engrained
into a variety of applications. In computer security, intrusion detection is an outstanding challenge. In the case of network intrusion
detection, data is collected at distributed sites and must be analyzed rapidly for signaling intrusion. The infeasibility of collecting this data
at a central location for analysis requires effective parallel and distributed algorithms. In the area of cryptography, some of the most
spectacular applications of Internet-based parallel computing have focused on factoring extremely large integers.

Embedded systems increasingly rely on distributed control algorithms for accomplishing a variety of tasks. A modern automobile consists
of tens of processors communicating to perform complex tasks for optimizing handling and performance. In such systems, traditional
parallel and distributed algorithms for leader selection, maximal independent set, etc., are frequently used.

While parallel computing has traditionally confined itself to platforms with well behaved compute and network elements in which faults
and errors do not play a significant role, there are valuable lessons that extend to computations on ad-hoc, mobile, or faulty
environments.

[ Team LiB ]
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks

[ Team LiB ]

1.3 Organization and Contents of the Text

This book provides a comprehensive and self-contained exposition of problem solving using parallel computers. Algorithms and metrics
focus on practical and portable models of parallel machines. Principles of algorithm design focus on desirable attributes of parallel
algorithms and techniques for achieving these in the contest of a large class of applications and architectures. Programming techniques
cover standard paradigms such as MPI and POSIX threads that are available across a range of parallel platforms.

Chapters in this book can be grouped into four main parts as illustrated in Figure 1.1. These parts are as follows:

Figure 1.1. Recommended sequence for reading the chapters.

This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks .

Fundamentals This section spans Chapters2 through 4 of the book. Chapter 2, Parallel Programming Platforms, discusses the physical
organization of parallel platforms. It establishes cost metrics that can be used for algorithm design. The objective of this chapter is not to
provide an exhaustive treatment of parallel architectures; rather, it aims to provide sufficient detail required to use these machines
efficiently. Chapter 3, Principles of Parallel Algorithm Design, addresses key factors that contribute to efficient parallel algorithms and
presents a suite of techniques that can be applied across a wide range of applications. Chapter 4, Basic Communication Operations,
presents a core set of operations that are used throughout the book for facilitating efficient data transfer in parallel algorithms. Finally,
Chapter 5, Analytical Modeling of Parallel Programs, deals with metrics for quantifying the performance of a parallel algorithm.

Parallel Programming This section includes Chapters6 and 7 of the book. Chapter 6, Programming Using the Message-Passing
Paradigm, focuses on the Message Passing Interface (MPI) for programming message passing platforms, including clusters. Chapter 7,
Programming Shared Address Space Platforms, deals with programming paradigms such as threads and directive based approaches.
Using paradigms such as POSIX threads and OpenMP, it describes various features necessary for programming shared-address-space
parallel machines. Both of these chapters illustrate various programming concepts using a variety of examples of parallel programs.

Non-numerical Algorithms Chapters 9–12 present parallel non-numerical algorithms. Chapter 9 addresses sorting algorithms such as
bitonic sort, bubble sort and its variants, quicksort, sample sort, and shellsort. Chapter 10 describes algorithms for various graph theory
problems such as minimum spanning tree, shortest paths, and connected components. Algorithms for sparse graphs are also discussed.
Chapter 11 addresses search-based methods such as branch-and-bound and heuristic search for combinatorial problems.Chapter 12
classifies and presents parallel formulations for a variety of dynamic programming algorithms.

Numerical Algorithms Chapters 8 and 13 present parallel numerical algorithms. Chapter 8 covers basic operations on dense matrices
such as matrix multiplication, matrix-vector multiplication, and Gaussian elimination. This chapter is included before non-numerical
algorithms, as the techniques for partitioning and assigning matrices to processors are common to many non-numerical algorithms.
Furthermore, matrix-vector and matrix-matrix multiplication algorithms form the kernels of many graph algorithms. Chapter 13 describes
algorithms for computing Fast Fourier Transforms.

[ Team LiB ]
This document was created by an unregistered ChmMagic, please go to http://www.bisenter.com to register it. Thanks .

[ Team LiB ]

1.4 Bibliographic Remarks

Many books discuss aspects of parallel processing at varying levels of detail. Hardware aspects of parallel computers have been
discussed extensively in several textbooks and monographs [CSG98, LW95, HX98, AG94, Fly95, AG94, Sto93, DeC89, HB84, RF89,
Sie85, Tab90, Tab91, WF84, Woo86]. A number of texts discuss paradigms and languages for programming parallel computersLB98, [
+
Pac98, GLS99, GSNL98, CDK 00, WA98, And91, BA82, Bab88, Ble90, Con89, CT92, Les93, Per87, Wal91]. Akl [Akl97], Cole [Col89],
Gibbons and Rytter [GR90], Foster [Fos95], Leighton [Lei92], Miller and Stout [MS96], and Quinn [Qui94] discuss various aspects of
parallel algorithm design and analysis. Buyya (Editor) [Buy99] and Pfister [Pfi98] discuss various aspects of parallel computing using
clusters. Jaja [Jaj92] covers parallel algorithms for the PRAM model of computation. Hillis H
[ il85, HS86] and Hatcher and Quinn H[ Q91]
discuss data-parallel programming. Agha [Agh86] discusses a model of concurrent computation based onactors. Sharp [Sha85]
addresses data-flow computing. Some books provide a general overview of topics in parallel computing [CL93, Fou94, Zom96, JGD87,
LER92, Mol93, Qui94]. Many books address parallel processing applications in numerical analysis and scientific computingDDSV99,
[
+
FJDS96, GO93, Car89]. Fox et al. [FJL 88] and Angus et al. [AFKW90] provide an application-oriented view of algorithm design for
problems in scientific computing. Bertsekas and Tsitsiklis [BT97] discuss parallel algorithms, with emphasis on numerical applications.

Akl and Lyons [AL93] discuss parallel algorithms in computational geometry. Ranka and Sahni RS90b] [ and Dew, Earnshaw, and
Heywood [DEH89] address parallel algorithms for use in computer vision. Green Gre91] [ covers parallel algorithms for graphics
applications. Many books address the use of parallel processing in artificial intelligence applications [Gup87, HD89b, KGK90, KKKS94,
Kow88, RZ89].

A useful collection of reviews, bibliographies and indexes has been put together by the Association for Computing Machinery [ACM91].
Messina and Murli [MM91] present a collection of papers on various aspects of the application and potential of parallel computing. The
scope of parallel processing and various aspects of US government support have also been discussed in National Science Foundation
reports [NSF91, GOV99].

A number of conferences address various aspects of parallel computing. A few important ones are the Supercomputing Conference,
ACM Symposium on Parallel Algorithms and Architectures, the International Conference on Parallel Processing, the International
Parallel and Distributed Processing Symposium, Parallel Computing, and the SIAM Conference on Parallel Processing. Important
journals in parallel processing include IEEE Transactions on Parallel and Distributed Systems, International Journal of Parallel
Programming, Journal of Parallel and Distributed Computing, Parallel Computing, IEEE Concurrency, and Parallel Processing Letters.
These proceedings and journals provide a rich source of information on the state of the art in parallel processing.

[ Team LiB ]
Random documents with unrelated
content Scribd suggests to you:
eastern borders of Ko-gu-ryŭ. Nominally Ko-gu-ryŭ held sway even
to the Japan Sea but practically the wild tribes roamed as yet at will
all through the eastern part of the peninsula. In the eighth year of
On-jo’s reign, 10 B.C., the Mal-gal forces besieged his capital and it
was only after a most desperate fight that they were driven back.
On-jo found it necessary to build the fortresses of Ma-su-sŭng and
Ch’il-chung-sŭng to guard against such inroads. At the same time
the Sŭn-bi were threatening Ko-gu-ryŭ on the north, but Gen. Pu
Bun-no lured them into an ambush and routed them completely. The
king rewarded him with land, horses and thirty pounds of gold, but
the last he refused.
The next year the wild men pulled down the fortresses lately
erected by King On-jo and the latter decided that he must find a
better site for his capital. So he moved it to the present site of Nam-
han, about twenty miles from the present Seoul. At the same time
he sent and informed the king of Ma-han that he had found it
necessary to move. The following year he enclosed the town in a
wall and set to work teaching agriculture to the people throughout
the valley of the Han River which flowed near by.
In the year which saw the birth of Christ the situation of affairs in
Korea was as follows. In the north, Ko-gu-ryŭ, a vigorous, warlike
kingdom, was making herself thoroughly feared by her neighbors; in
the central western portion was the little kingdom of Păk-je, as yet
without any claims to independence but waiting patiently for the
power of Ma-han so to decline as to make it possible to play the
serpent in the bosom as Wi-man had done to Ki-ja’s kingdom. In the
south was Sil-la, known as a peaceful power, not needing the sword
because her rule was so mild and just that people from far and near
flocked to her borders and craved to become her citizens. It is one
of the compensations of history that Sil-la, the least martial of them
all, in an age when force seemed the only arbiter, should have finally
overcome them all and imposed upon them her laws and her
language.
 Chapter VII.
Change of Ko-gu-ryu capital.... Sil-la raided.... Legend of Suk-ta’l-ba.... fall of Ma-
han.... beginning of Chinese enmity against Ko-gu-ryu....the three kingdoms
differentiated.... King Yu-ri degraded.... extension of Ko-gu-ryu.... Japanese
corsairs... remnant of Ma-han revolts.... fall of Pu-yu.... origin of in-gum....
siege of Ko-gu-ryu capital raised.... Sil-la’s peaceful policy.... patronymics....
official grades.... unoccupied territory.... kingdom of Ka-rak.... legends....
position.... dependencies.

We read that in 2 A.D. the king of Ko-gu-ryŭ was about to sacrifice

a pig to his gods, when the pig escaped and taking to its heels was
chased by the courtier Sŭl-chi into the district of Kung-nă. He caught
the animal near Wi-na Cliff, north of the Ch’o-san of today. When he
returned he described the place to the king as being rough and
consequently suitable for the site of a capital. Deer, fish and turtles
also abounded. He gave such a glowing account that the king was
fain to move his capital to that place, where it remained for two
hundred and six years.
In 4 A.D. Hyuk-kŭ-se, the wise king of Sil-la died and seven days
later his queen followed him. It is said that they were so completely
one that neither could live without the other. Nam-hă his son, with
the title of Ch’a-ch’a-ung, reigned in his stead. A remnant of the
Nang-nang tribe, hearing of the death of King Hyŭk-kŭ-se, thought it
a fitting time to make a raid into Sil-la territory, but they were
beaten back.
In the third year of his reign, Nam-hă built a shrine to his father
and then put the management of the government into the hands of
a man named Sŭk-t’al-hă who had become his son-in-law. This man
is one of the noted men of Sil-la and his origin and rise are among
the cherished traditions of the people.
 Somewhere in north-eastern Japan there was a kingdom known as
Ta-p’a-ra and there a woman, pregnant for seven years, brought
forth an egg. The neighbors thought it a bad omen and were minded
to destroy it but the mother, aware of their intentions, wrapped the
egg in silk and cotton and placing it in a strong chest committed it to
the waters of the Japan Sea. In time it drifted to A-jin Harbor on the
coast of Sil-la where an old fisherwoman drew it ashore and found
upon opening it that it contained a beautiful child. She adopted him
and reared him in her humble home. It was noticed that wherever
the child went the magpies followed him in flocks, so they gave him
the name of Sŭk, the first part of the Chinese word for magpie. The
second part of his name was T’al, “to put off” referring to his having
broken forth from the egg, and the final syllable of his name was Hă
meaning “to open” for the fishwife opened the chest. This boy
developed into a giant both physically and mentally. His foster-
mother saw in him the making of a great man, and so gave him
what educational advantages she could afford. When he had
exhausted these she sent him to enter the service of the great
statesman Pyo-gong the same that had acted as envoy to Păk-je.
Pyo-gong recognised his merit and introduced him at court where
his rise was so rapid that ere long he married the king’s daughter
and became vicegerent of the realm, the king resigning into his
hands the greater part of the business of state.
The year 9 A.D. beheld the fall of the kingdom of Ma-han. We
remember that Ki-jun became king of Ma-han in 193 B.C. He died
the same year and was succeeded by his son Ki-t’ak with the title
Kang-wang, who ruled four years. It was in 58 B.C. that Ki-jun’s
descendant Ki-hun (Wun-wang) ascended the throne. It was in the
second year of his reign that Sil-la was founded and in his twenty-
second year that Ko-gu-ryŭ was founded. After twenty-six years of
rule he died and left his son, Ki-jŭng, to hold the scepter. It was this
king who, in his sixteenth year gave On-jo the plot of land which
became the seat of the kingdom of Păk-je. Twenty-six years had now
passed since that act of generosity. Păk-je had steadily been growing
stronger and Ma-han had as steadily dwindled, holding now only the
two important towns of Wŭn-san and Köm-hyŭn. In fact some
authorities say that Ma-han actually came to an end in 16 B.C. at the
age of 177 years but that a remnant still held the towns of Wŭn-san
and Köm-hyŭn. The balance of proof is however with the statement
that Ma-han kept up at least a semblance of a state until 9 A.D.
The first sign of hostile intent on the part of Păk-je against her
host, Ma-han, had appeared some years before, when Păk-je had
thrown up a line of breast-works between herself and the capital of
Ma-han. The latter had no intention of taking the offensive but Păk-
je apparently feared that Ma-han would divine her hostile intent. Ma-
han hastened to send a message saying “Did I not give you a
hundred li of land? Why do you then suspect me of hostile designs?”
In answer, Păk-je partly from shame and partly because she saw
that Ma-han was wholly unsuspicious of her ulterior designs, tore
down the barriers and things went on as before. But now that Ma-
han was utterly weak, the king of Păk-je decided to settle the matter
by one bold stroke. He organised a great hunting expedition and
under cover of this approached the Ma-han capital and took it
almost without resistance. Thus, as Wi-man had paid back the
kindness of Ki-jun by treachery so now again On-jo paid back this
last descendant of Ki-jun in the same way.
Up to this time China had looked on with complacency at the
growth of Ko-gu-ryŭ but now Wang-mang the usurper had seized
the throne of the Han dynasty. His title was Hsin Whang-ti. One of
his first acts seems to have been directed against the powerful little
kingdom that had supplanted the two provinces of Tong-bu and
P‘yŭng-ju into which China had divided northern Korea. He was
probably suspicious of a rapidly growing and thoroughly warlike
power which might at any time gather to its standards the wild
hordes of the north and sweep down into China.
Here was the beginning of a long struggle which lasted with
occasional intermissions until Ko-gu-ryŭ was finally destroyed some
eight centuries later. Ko-gu-ryŭ was uniformly China’s foe and Sil-la
was as uniformly her friend and ally. Păk-je was now one and now
the other. It may be in place to say here that the three powers that
divided the peninsula between them were strongly differentiated.
Ko-gu-ryŭ in the north was a strong, energetic, fierce, unscrupulous
military power, the natural product of her constituent elements. Sil-la
was the very opposite; always inclined toward peace and willing
oftentimes to make very large concessions in order to secure it. Her
policy was always to conciliate, and it was for this mainly that at the
last China chose her as the one to assume control of the whole
peninsula. Păk-je differed from both the others. She was as warlike
as Ko-gu-ryŭ but as weak in military resources as Sil-la. She
therefore found her life one scene of turmoil and strife and she was
the first of the three to succumb.
It was in 12 A.D. that Wang-mang sent an envoy to Yu-ri, king of
Ko-gu-ryŭ, demanding aid in the work of subduing the wild tribes of
the north. This was refused by the headstrong Yu-ri, but the
Emperor compelled him to send certain troops to accompany the
Chinese army. They however took advantage of every opportunity to
desert, and large numbers of them formed a marauding band that
penetrated the Liao-tung territory and plundered and killed on every
hand. For this cause the Emperor sent against Ko-gu-ryŭ a strong
force under Gen. Om-u, who speedily brought the recalcitrant Yu-ri
to terms, took away his title of royalty and left him only the lesser
title of Hu or “Marquis.” From that day began the policy of reprisals
on Chinese territory which Ko-gu-ryŭ steadily pursued until it cost
her life.
These were stirring days in all three of the kingdoms of the
peninsula. In 14 A.D. Ko-gu-ryŭ extended her territory northward by
the conquest of the Yang-măk tribe and at the same time she seized
a strip of land beyond the Liao River This shows that the castigation
inflicted by Wang-mang had not been very severe.
At the same time Sil-la was being harrassed along her southern
sea-board by Japanese corsairs, and while her small army was busy
driving these out the wild people of Nang-nang attacked her on the
north. It is said that one night a meteor fell in their camp and
frightened them back to their own country and thus Sil-la was saved.
Two years later the king Yu-ri of Ko-gu-ryŭ died and his son Mu-
hyŭl ascended the throne, bestowing on his father the title Tong-
myŭng or “Eastern Brightness.” The same year saw a remnant of the
overthrown kingdom of Ma-han, under the leadership of Captain
Chu-geun, attempt to wrest the scepter from Păk-je and restore the
fallen house, but they were defeated and together with their wives
and children were put to the sword. About this time an ancient royal
seal was unearthed in northern Sil-la, where Kang-neung now lies. It
became the royal seal of Sil-la.
The next year Ko-gu-ryŭ, ever on the lookout for aggrandisement,
made the conquest of Pu-yŭ, the land from which Chu-mong had
fled. The tradition is as follows. Ta-so the king of Pu-yŭ, had become
possessed of a red crow with two bodies but only one head. The
soothsayers said “Two countries will be joined under one head”. The
king replied “Then it means that I shall conquer Ko-gu-ryŭ.” So he
sent the bird to the king of Ko-gu-ryŭ as a gage of war, but that
astute monarch replied “Red is the color of the south. I shall
therefore conquer you.” Thereupon he took the initiative and sent a
powerful army northward to make good his threat. The story says
that as the army entered Li-mul forest the soldiers found swords
clashing together but wielded by invisible hands. These they seized
and hastened on. Soon they were joined by a gigantic warrior with a
white face who joined their party and gave his name as Kwe-yu.
Approaching the capital of Pu-yŭ, they brought up at night before
an extensive marsh. The Pu-yŭ king, thinking to surprise them by a
night attack, attempted to cross the marsh, but became mired. The
giant Kwe-yu dashed into the swamp and brought to the Ko-gu-ryŭ
king his rival’s head. Upon this the Pu-yŭ forces surrendered; all but
the brother of the fallen king who fled with a hundred followers and
settled near the Ya-lu River, calling the place Kal-sa. This Ko-gu-ryŭ
winked at.
In 24 A.D. the king of Sil-la died, having nominated as his
successor not his son but Sŭk-t’al-hă his son-in-law. After the
obsequies had been performed Sŭk-t’al-hă insisted that the prince
assume the throne, but he in turn insisted that the dead king’s
orders be followed. As a compromise Sŭk-t’al-hă proposed that they
should find a man with sixteen teeth in his upper jaw, as this was a
sign of unusual wisdom, and that upon him the throne should be
bestowed. When it came to the test, it was found that the prince
himself was the man. He could no longer refuse and ascended the
throne under the title of Yi-sa-geum, or “Sixteen Teeth.” The present
word In-gum which means “king” was doubtless derived from or is a
corruption of this Sil-la word.
Meanwhile Ko-gu-ryŭ had been pushing her conquests steadily.
Kă-ma and Ku-da, two northern districts or “kingdoms” were
absorbed and other conquests were contemplated. The Emperor
beheld these enlargements of Ko-gu-ryŭ with some concern and in
27 A.D. sent a strong force to bring her to terms. At the first
encounter the forces of Ko-gu-ryŭ were routed and fell back toward
the capital which, as we have seen, was then at or near the present
town of Eui-ju. The king hastily summoned a council of war at which
it was decided to man the walls of the capital and try to hold out
until the enemy should be compelled by lack of food or the severity
of the weather to raise the siege. The Chinese knew that there was
little water within the wall and had high hopes of compelling a
speedy surrender. This was all too true and there was soon much
distress in the city; but a certain courtier said “If you will give me all
the fish in the city I will undertake to make the enemy raise the
siege or I will pay the penalty with my life.” He was given permission
and soon he had the soldiers along the wall going through the
motions of a bath, using fish scales for water. The scales glittered in
the sun like drops of water and the enemy supposing that there
must therefore be a good store of water in the city despaired of
taking it by siege and so struck their tents and returned to China.
The marked difference between Ko-gu-ryŭ and Sil-la was well
illustrated by the events of this year. While Ko-gu-ryŭ was reaching
out covetous hands in every direction and carrying fire and sword
into the hamlets of inoffensive neighbors, Sil-la was pursuing a
course of such good will to all both without and within her borders
that natives of the wild tribes to the north of her came in large
numbers and settled on her soil, glad to become citizens of so kind
and generous a land. The king himself made frequent tours of the
country alleviating the distress of widows, orphans and cripples. It
was in 32 B.C. that he changed the name of the six original families
which united in founding Sil-la. The men of Yang-san, Ko-hŭ, Tă-su,
Ul-jin, Ka-ri, and of Myŭng-whal were named respectively Yi, Ch’oé,
Son, Chöng, Pă and Sŭl. These names will be recognised at once as
among the most common patronymics in Korea at the present day,
which adds confirmatory evidence that Korea of to-day is essentially
the Korea of the south. When we add to this the fact that the names
Pak, Kim, An, Ko, Sŭk, Yang, So, Sŭ, Kwŭn, Pă, Im, Na, Hyŭn, Kwak,
Ho, Whang, Chang, Sim and Yu originated in southern Korea the
argument becomes well-nigh conclusive. The only names of
importance that did not originate in southern Korea are Min, Song,
Om, Cho, and Han; and many of these originated in what must have
been Ma-han territory. At the same time the king established
seventeen official grades and called them respectively I-bŭl-son, I-
ch’ŭk-son, I-son, P’a-jin-son, Tă-a-son, A-son, Kil-son, Sa-son, etc.
It must be remembered that as yet neither of the “Three
Kingdoms” had begun to occupy all the territory that nominally
belonged to it or that lay within its “sphere of influence.” Between
them lay large tracts of land as yet unoccupied except by wild tribes.
It is more than probable that at no point did any of these kingdoms
actually touch each other. Ko-gu-ryŭ was broadening out
northwards, Păk-je was at a standstill and Sil-la was growing rather
by immigration than by occupation of new territory. As yet Sil-la had
taken but four districts outside of the original six, and so we see that
a large part of the south was still in the hands of the original
inhabitants as given in the list of the settlements of the three Hans.
In 41 A.D. the nine districts whose names ended in kan, namely A-
do-gan, Yö-do-gan, P’i-do-gan, O-do-gan, Yu-su-gan, Yu-ch’ŭn-gan,
Sin-ch’ŭn-gan, Sin-gwi-gan and O-ch’ŭn-gan, formed a confederacy
and called it the “Kingdom of Ka-rak”. They placed their capital at
Ka-rak, the present town of Kim-hă, and made Keum Su-ro their
king. Tradition says that he obtained his Queen in the following way.
A boat approached the shore bearing a beautiful woman, Queen Ho,
whose ornamental name was Whang-ok or “Yellow Jade”. She came
from the far southern kingdom of A-yu-t’a, otherwise known as
Ch’ŭn-ch’uk. It is said that she lived a hundred and fifty-seven years
and that the king survived her one year. All that is told us of the
history of this rival of Sil-la is the list of her kings which will be found
in the chronological tables. After an existence of 491 years it came
to an end in the reign of the Sil-la king Pŭp-heung. It is also affirmed
that when Sil-la fell in 935, some worthless wretches who defiled the
grave of Keum Su-ro were mysteriously killed, one by the falling of a
beam, one by an invisible archer and nine others by a serpent
eighteen feet long. The records say that when the Japanese, at the
time of the great invasion three centuries ago, dug open this king’s
grave they found great store of gold and jade. The skull of the
monarch was of prodigious size, and beside his body lay two women
whose features were well preserved but which dissolved and melted
away when exposed to the air. It is barely possible that we here
have an indication that embalming was practiced, but if so we have
no other intimation of it.
Ka-rak extended eastward as far as Wang-san River, six miles to
the west of the present Yang-san; to the north-east as far as Ka-ya
San, the present Ko-ryŭng; to the south and south-west as far as the
coast and on the west to Chi-ri San. From this we see that it was
little inferior to Sil-la in size.
Ka-rak had five dependencies, namely the districts known under
the common name of Ka-ya. They were So-ga-ya, Ko-ryŭng-ga-ya,
Song-san-ga-ya, Tă-ga-ya and A-ra-ga-ya. They correspond
respectively to the present towns of Ko-sŭng, Ham-ch’ang, Ham-
ch’ang, Sŭng-ju, Ko-ryŭng and Ham-an. Tradition says that one day
when the chiefs of the nine tribes of Ka-rak were banqueting they
saw upon the slope of Sung-bong, called also Ku-yii-bong, a singular
cloud. From the sky above it came a voice. They hastened up the
mountain and there found a golden box containing six golden eggs.
These opened and disclosed six boys. One of them was Keum Su-ro
who became king of Ka-rak and the other five were made chiefs of
the five Ka-ya, subject to Ka-rak. Of these Ka-ya states we know the
founder of only one. He was descended from Kyŏn-mo-ju, the
female divinity of Ka-ya Mountain who wedded a celestial being, Yi-
ja-ga. Their off-spring was Yi-i-a-si, who founded one of the Ka-ya
states. The Ka-ya states fell before Sil-la some five hundred years
later in the reign of King Chin-heung.
 Chapter VIII.
Vicissitudes of Ko-gu-ryu.... last Ma-han chief joins Sil-la.... Pak-je and Sil-la
become sworn enemies.... legend of Kye-rim.... Pak-je worsted.... Ko-gu-ryu’s
strength on the increase.... Sil-la’s rapid growth.... Ka-ya attacks Sil-la.... Ko-
gu-ryu make compact with Ye-mak.... Su-sŭng’s evil reign.... roads in Sil-la....
Japanese raid.... legend.... an epicurean.... Pak-je’s victory.... origin of
government loans.... Yun-u’s trickery.... capital of Ko-gu-ryu moved.... wild
tribes attack Sil-la.... democratic ideas in Sil-la.... Ko-gu-ryu breaks with
China.... and attacks Sil-la.... China invades Ko-gu-ryu.... the king retreats....
relieved through treachery.... capital of Ko-gu-ryu moved to P’yung-yang....
beginning of feud between Korea and Japan.... reforms in Pak-je.... third
century closes.... progress of Sil-la.... how Eul-bul became king of Ko-gu-
ryu.... a noble lady of Sil-la is sent to Japan.

Mu-hyŭl, the third king of Ko-gu-ryŭ died in 45, leaving the

kingdom to the tender mercies of his son a worthless debauchee.
Four years later he in turn made way for Hă-u, a member of a
collateral branch of the family. Following the traditions of Ko-gu-ryŭ
this ruler professed loyalty to China on the one hand and seized all
the Chinese territory he could lay hands on, on the other. In 54 he
was assassinated by one Tu-no and the seven year old grandson of
king Yu-ri was placed on the throne, a regent being appointed to
carry on the government until the boy reached his majority. The
good work continued. Ten forts were built in western Liao-tung to
guard against Chinese advances, which shows that she had regained
nearly all the territory she had lost at the hands of the parvenu
Wang-mang. The following year she took formal possession of the
territory of Ok-jŭ on the eastern coast.
In the year 58 Yu-ri, the third king of Sil-la died. He must not be
confounded with Yu-ri the second king of Ko-gu-ryŭ. The sound is
the same but the character is different. It was he who had the
difference of opinion with Sŭk-t’al-hă in regard to the succession. As
he died without issue the reins of government naturally passed into
the hands of the aged statesman Sŭk-t’al-hă. He was sixty-two years
old when he assumed the cares of royalty. In his fifth year the one
remaining Ma-han chief, Măng-so, who had escaped the appetite of
Păk-je, went over to Sil-la, as he concluded it was no longer possible
to prolong a hopeless struggle against Păk-je. Pok-am fortress thus
passed into the hands of Sil-la. Strange to say Păk-je not only did
not resent this but even made overtures to Sil-la for a friendly
meeting of their respective kings in the following year. Sil-la refused
to sanction this, and the rebuff was too much for the equanimity of
Păk-je. From that day the attitude of Păk-je toward Sil-la was one of
studied hostility, broken only by an occasional spasmodic attempt at
reconciliation. Among the three kingdoms, Sil-la was the only one
that preserved her dignity intact and kept herself untainted by the
charge either of avarice or pusilanimity.
The year 66 brought forth another of those wonders that
embellish the legendary lore of Korea. The king of Sil-la was
wakened one night by the loud cackling of a hen, which seemed to
come from a forest to the south. A messenger was sent to see what
was the cause of the disturbance and he found a box hanging from
the branch of a tree, while on the ground beneath it there cluttered
a white hen. When the box was placed before the king and he had
opened it a handsome child was found. It received the name Keum
Yun-ji. Some say this Yŭn-ji was merely a part of the name while
others affirm that it is a pure Sil-la word meaning “baby”. Up to this
time the kingdom had been called Sŭ-ra-bŭl but now the king
changed it to Kye-rim, kye meaning “hen” and rim meaning “woods.”
So the kingdom was called “Hen in the Woods”, not a very dignified
name but one, perhaps, that fitted well the military prowess of the
kingdom.
In 68 Păk-je deemed herself strong enough to undertake
operations against Sil-la. She began by seizing the fortress of Wa-
san. She enjoyed possession of it for nine years but in the end she
paid dear, for it was retaken by Sil-la and the Păk-je garrison was put
to the sword. This year also saw a continuation of Ko-gu-ryŭ’s
forward policy and the little settlement of Kal-sa which had been
make by Pu-yŭ fugitives was absorbed. She followed this up by the
conquest of Chu-ra farther north. Her military strength seems to
have been on the rapid increase.
In 80 the great Sŭk-t’al-hă died and was succeeded by the son of
King Nam-hă. He must have been of advanced age and yet not so
old as to prevent his becoming the greatest conqueror that Sil-la
ever produced. During the thirty-two years of his reign he added to
the Sil-la crown the districts of Eum-jip-pŭl, Ap-to, Pi-ji, Ta-bŭl, Ch’o-
p’al, and Sil-jik. These together with U-si and Kŭ-ch’il, which and
been added the year before his accession, formed a considerable
increase in the territory of the kingdom and added not a little to Sil-
la’s reputation as a military power. This king, P’a-sa, was one of
those men who seem to take hold of affairs by the right end and
wring success from seeming failure. He was as great an
administrator as he was mild a conqueror. He attended so carefully
to the needs of the people that it is said that during most of his
reign food was so plentiful that the wayfarer needed no money to
pay for food or lodgings along the road.
The kingdom of Ka-ya, whose origin we noted in the previous
chapter, now assumed the offensive against Sil-la. The first
intimation we have of this is the fact that Sil-la in 88 built two forts
named Ka-so and Ma-du, the first of which was to guard against the
encroachments of Păk-je and the second to guard against those of
Ka-ya. It was not till three years later that Ka-ya actually opened
hostilities by inaugurating an expedition against Sil-la. As the event
is not disclosed by the annalists we may conclude that it was
unsuccessful.
Ko-gu-ryŭ now extended the field of her military operations. She
made friends with the people of Ye-măk, to the east, and together
with them began a series of raids into Chinese territory beyond the
northern borders. The sixth king of Ko-gu-ryŭ, T’ă-jo Wang, had now
reached the sixty-ninth year of his reign so he turned over to his
brother, Su-sŭng, the administration of affairs. This brother was as
ambitious as the king and continued the league with Ye-măk and the
encroachments upon China. But he was disloyal to his brother and
tried to form a combination against him. In this he was not
successful. The reign of this T’ă-jo Wang was the longest one on
record in Korean annals. He held the scepter ninety-four years,
thereby sorely trying the patience of his heir apparent. That
gentleman came to the throne at the green old age of seventy-six, in
the year 147 A.D. He showed however that his memory had not yet
failed him for one of his first acts was to arrest and put to death all
the wise men who had chidden him for attempting to unseat his
brother. Ko Pok-chang a celebrated scholar of that day was so
overwhelmed in view of this barbarous act that he asked to be
destroyed with the rest of the wise men, a wish that was probably
granted. One day this singular monarch having seen a white fox
cross his path, an evil omen, asked a soothsayer what it might
portend. That individual suggested that if the king should reform
even the worst of omens would turn out happily. The soothsayer lost
his head as a result of his candor; but from that day on, whenever
the king wanted to consult a soothsayer he found that they were all
engaged in important work at some distant point.
King Il-seung of Sil-la whose reign began 134 was the first to pay
attention to the building of good roads throughout the country. In
his fifth year he built a road from his capital to Chuk-yŭn, now Pung-
geui, and another one over Kye-ip Pass. These became very
important thoroughfares. We also find that his successor continued
this good work by opening roads thro to the north of the kingdom.
These kings were not many years behind the Romans in recognising
the vast importance of good roads both for administrative and
military purposes.
The relations between Sil-la and Japan are graphically described in
the single statement that when someone circulated in the capital the
rumor that a company of Japanese were coming the people fled
precipitately from the city until it was half depopulated. When the
mistake was discovered they gradually came back.
The interesting legend of Yŭng-o and Se-o belongs to the year
158, though it scarcely merits the “once upon a time” of a nursery
tale. Yŭng-o a poor fisherman lived with his wife Se-o beside the
waters of the Japan Sea on the eastern shore of Sil-la. One day as
Yŭng-o was seated on a great boulder beside the water, fishing, he
felt the rock tremble and then rise straight in air. He was carried, to
his great consternation, eastward across the sea and deposited in a
Japanese village. The Japanese folk took him for a god and made
him their king at once. When his wife found that he did not return
from fishing she went in search of him. Ascending the same rock
that had carried him to Japan she experienced the same novel
extradition that had so surprised her spouse. She found him
metamorphosed into a king and was nothing loath to become queen.
But their departure brought disaster to Sil-la for the sun and moon
were darkened and the land was shrouded in gloom. The sooth-
sayers said it was because someone had gone to Japan. An envoy
was sent post haste to those islands in search of the fugitives, but
found to his dismay that they had become king and queen of one of
the kingdoms there. He told his story and besought them to return,
but they seemed well satisfied with the change. Se-o however
brought out a roll of silk and gave it to the envoy saying that if the
king of Sil-la would spread it out and sacrifice upon it the light would
return. The event proved the truth of her statement and when the
king uttered the words of invocation the sunlight burst forth again
and all was well. It is an interesting but melancholy fact that most of
the arguments used to show a Korean origin of things Japanese are
based upon evidence nearly if not quite as credible as this story. The
Japanese work entitled the Kojiki bears the same relation to the
carefully detailed history of Sil-la that the Niebelungenlied bears to
the works of Tacitus.
When the time came for Su-sŭng, the sanguinary king of Ko-gu-
ryŭ to die a young scapegrace by the name of Ch’a-da came to the
throne. His idea of royalty was that it consisted in one long orgie. He
attempted to carry out his ideal but was cut short within a year by
the assassin’s knife. His motto, in his own words, was “Who does not
wish to enjoy life?” Epicureanism may have existed in Korea before
but it had never had so frank a disciple. Păk-ko a relative of the
murdered king was called from a mountain fastness whither he had
fled for safety. They had to ask him three times before they could
convince him that it was not a mere decoy.
 By the year 168 either Păk-je had grown so strong or Sil-la so
weak that the former deemed it a fit time to make a grand
demonstration all along Sil-la’s western border. It is said she carried
back a thousand captives to grace her triumph. Sil-la, though filled
with rage, was not in condition to return the compliment in kind. She
however sent an urgent letter pointing out the advantages of peace
and asking that the captives be returned. We may imagine how this
was received by the proud army flushed as it must have been by an
unwonted victory.
About this time was begun one of the ancient customs of Korea
that has ever since exerted an important influence upon the life of
the people. While hunting the king met a man weeping bitterly and
upon being asked what was the matter replied that he had not a
grain of food to give his parents. Thereupon the king gave him an
order on the government granary with the understanding that when
autumn came he should pay it back. Thus originated the whan-sang
or custom of making government loans in the spring to be paid back
with interest in the autumn. When this king died he was succeeded
by the grandson of old Sŭk-t’al-hă. He took in hand the work of
instilling new life into the well-nigh dead bones of Sil-la. His first
action was to establish two military stations at the capital so that it
might not be at the mercy of the first adventurer that might pass
that way. He also ordered the people to pay less attention to the
construction of fine government buildings and more to agriculture,
the back bone of the state.
Nam-mu the tenth king of Ko-gu-ryŭ died at night and the queen,
desiring to gain an extension of her power, slipped out of the palace
and hastened to the house of the king’s oldest brother Pal-gi. She
stated the case and urged him to hasten to the palace and assume
the royal prerogative. He refused to believe that the king was dead
and accused her of immodesty. She then hurried to the house of the
younger brother Yŭn-u and repeated the story. The young man
accompanied her and when morning broke it was found that he was
established in the palace and ready to meet all comers. Pal-gi raged
and cursed. He stormed the palace with his retainers, but being
unsuccessful, was fain to beat a retreat to Liao-tung.
 The dawn of the third century saw the three states of Korea in the
same relative position as before. Ko-gu-ryŭ was still the same
ambitious military power, Păk-je was still her own worst enemy
though flaunting for the time being in the gay colors of a temporal
triumph, Sil-la was plodding along quietly paying more attention to
internal improvements and so earning the right which she afterward
enjoyed of holding sway over the whole peninsula. The first twenty-
five years of the century witnessed unusual activity on the part of
the surrounding savages who in view of the constantly increasing
power of the three states beheld their territories diminishing. The
wild people of Kol-p’o, Chil-p’o and Ko-p’o ravaged the borders of Sil-
la but were driven back. On the south she attacked and burned a
settlement of Japanese corsairs who had apparently gained a
foothold on the mainland. Păk-je was also attacked on the east by
the savages and was obliged to build a wall at Sa-do to keep them
back. This period saw over a thousand Chinese refugees cross the
Yalu and find asylum in Ko-gu-ryŭ. It also saw U-wi-gŭ, the fruit of a
liaison between the eleventh king of Ko-gu-ryŭ and a farmer girl
whom he met while hunting, ascend the throne of Ko-gu-ryŭ. It
witnessed a remarkable exhibition of democratic feeling in Sil-la
when the people rejected Prince Sa-ba-ni and in his place set up Ko-
i-rŭ to be king.
The year 240 was an important one in the history of Ko-gu-ryŭ.
King U-wi-gŭ was a man of boundless ambition and his temerity was
as great as his ambition. Ko-gu-ryŭ had been at peace with China for
eight years when, without warning, this U-wi-gŭ saw fit to cross the
border and invade the territory of his powerful neighbor. The town of
An-p’yŭng-hyŭn in western Liao-tung fell before the unexpected
assault. This unprovoked insult aroused the slumbering giant of the
Middle Kingdom and the hereditary feud that had existed for many
years between Ko-gu-ryŭ and China was intensified. At the same
time U-wi-gŭ turned his eyes southward and contemplated the
subjugation of Sil-la. To this end he sent an expedition against her in
the following year. It was met on the Sil-la border by a defensive
force under Gen. Sŭk U-ro who withstood the invaders bravely but
was driven back as far as the “Palisades of Ma-du” where he took a
firm stand. As he could not be dislodged the invading army found
itself checked. Meanwhile a dark cloud was rapidly overspreading
Ko-gu-ryŭ’s western horizon. The great Chinese general, Mo Gu-
geum, with a force of 10,000 men advanced upon the Ko-gu-ryŭ
outposts and penetrated the country as far as the present Sŭng-
ch’ŭn where he met the Ko-gu-ryŭ army under the direct command
of king U-wi-gŭ. The result was an overwhelming victory for Ko-gu-
ryŭ whose soldiers chased the flying columns of the enemy to Yang-
băk-kok where dreadful carnage ensued. “Whom the gods would
destroy they first make mad” proved true in this case. U-wi-gŭ was
so elated over the victory that he declared that a handful of Ko-gu-
ryŭ troops could chase an army of Chinese. Taking five hundred
picked cavalry he continued the pursuit; but he had boasted too
soon. Gen. Mo Gu-geum’s reputation was at stake. Rallying a handful
of his braves the latter turned upon his pursuers and handled them
so severely that they turned and fled. The Chinese followed up the
timely victory and threw themselves upon the army of Ko-gu-ryŭ so
fiercely that the tables were completely turned. It is said that in the
engagement that followed Ko-gu-ryŭ lost 18,000 men. King U-wi-gŭ,
seeing that all was lost, fled back to his capital and awaited
developments. But Gen. Wang-geui, Mo Gu-geum’s associate,
pursued the king across the Yalu and gave him no rest until he had
fled eastward to the territory of Ok-jŭ on the eastern coast. On his
way thither he crossed Chuk-nyŭng Pass where all his remaining
guard forsook him and fled. One of his officials, Mil-u, said “I will go
back and hold the enemy at bay while you make good your escape”.
So with three or four soldiers he held the narrow pass while the king
found a retreat in a deep valley, where he succeeded in getting
together a little band of soldiers. He offered a reward to anyone who
should go and bring Mil-u safely to him. U Ok-ku volunteered to go.
Finding Mil-u wounded and lying on the ground he took him in his
arms and carried him to the king. The latter was so delighted to
recover his faithful follower that he nursed him back to life by his
own hand. A few days later the pursuit continued and the king was
again hard pressed. A courtier, Yu-ryu, offered to go to the enemy’s
camp and in some way stop the pursuit. Taking some food he went
and boldly announced that the king desired to surrender and had
sent this gift ahead to announce his coming. His words were
believed and the general received the gift. But Yu-ryu had concealed
a short sword beneath the dishes and when he approached the
general he whipped out the weapon and plunged it into the enemy’s
breast. The next moment he himself was cut down by the
attendants. When the king learned that the pursuers had lost their
general he rallied his little force, threw himself upon them and put
them to flight. The following year U-wi-gŭ, recognising that his
capital was too near the border, decided to remove the court to
P‘yŭng-yang which had been the capital for so many centuries. Two
years later he made a treaty with Sil-la which remained unbroken for
a century. He had been cured of some of his over-ambitiousness.
Yŭn-bul was his successor.
In the third year of King Ch’ŭm-hă of Sil-la, 249 A.D. the first
envoy ever received from Japan arrived at the shore of Sil-la. He was
met by Gen. Sŭk U-ro who addressed him in the following
unaccountable manner, “It would be well if your king and queen
should come and be slaves in the kitchen of the king of Sil-la”.
Without a word the envoy turned about and posted back to Japan.
An invasion of Korea was determined upon and soon a powerful
force landed on the coast of that country. Gen. Sŭk U-ro was filled
with dismay and remorse. He confessed to the king that he was the
cause of this hostile display and begged to be allowed to go alone
and propitiate the advancing enemy. It was granted and he walked
straight into the Japanese camp and confessed his crime and asked
that he alone be punished. The Japanese took him at his word,
burned him alive in their camp and returned to their own land
without striking a blow. The following year the same envoy came
again and was well received by the king, but the widow of Gen. Sŭk
U-ro desiring to avenge the blood of her husband, obtained
permission to work in the kitchen of the envoy’s place of
entertainment. There she found opportunity to poison his food and
thus accomplish her purpose. This of course put an end to all hope
of amity between the two countries and that event marks the
beginning of the feud which in spite of occasional periods of
apparent friendship, existed between the people of Japan and Korea
until the year 1868. Hostilities did not however begin at once.
The latter half of the third century beheld few events of special
interest in the peninsula. During this period Păk-je seems to have
made a spasmodic effort at reform, for we read that she reorganised
her official system and set a heavy penalty for bribery, namely
imprisonment for life. She also patched up a shallow peace with Sil-
la. In Ko-gu-ryŭ a concubine of King Pong-sang tried to incense him
against the queen by showing him a leathern bag which she claimed
the queen had made to drown her in. The king saw through the trick
and to punish the crafty concubine had her killed in the very way she
had described. A chief of the Sŭn-bi tribe invaded Ko-gu-ryŭ and
desecrated the grave of the king’s father. The wild men of Suk-sin
attempted to overthrow Sil-la but the king’s brother drove them back
and succeeded in attaching their territory to the crown of Sil-la. It is
said that when Sil-la was hard pressed by a band of savages strange
warriors suddenly appeared and after putting the savages to flight,
as suddenly disappeared. Each of these strange warriors had ears
like the leaves of the bamboo and when it was discovered next day
that the ground around the king’s father’s grave was covered with
bamboo leaves it was believed that he had come forth from his
grave with spirit warriors to aid his son.
With the opening of the fourth century the fifteenth king of Sil-la,
Ki-rim, made an extensive tour of his realm. He passed northward as
far as U-du-ju near the present Ch’un-ch’ŭn. He also visited a little
independent “kingdom” called Pi-ryul, now An-byŭn, and made many
presents, encouraged agriculture and made himself generally
agreeable. Not so with the king of Ko-gu-ryŭ. He was made of
sterner stuff. He issued a proclamation that every man woman and
child above fifteen years old should lend their aid in building a
palace. Ko-gu-ryŭ had of late years passed through troublous times
and the people were in no mood to undertake such a work. An
influential courtier, Ch’ang Cho-ri, attempted to dissuade the king
but as he was not successful he settled the question by
assassinating the king. Eul-bul, who succeeded him, had a
chequered career before coming to the throne. Being the king’s
cousin he had to flee for his life. He first became a common coolie in
the house of one Eun-mo in the town of Sil-la. By day he cut wood
on the hill sides and by night he made tiles or kept the frogs from
croaking while his master slept. Tiring of this he attached himself to
a salt merchant but being wrongfully accused he was dragged before
the magistrate and beaten almost to death. The official Ch’ang Cho-
ri and a few others knew his whereabouts and, hunting him up, they
brought him to the “Pul-yu water” a hundred and ten li from P’yŭng-
yang, and hid him in the house of one O Măk-nam. When all was
ripe for the final move, Ch’ang Cho-ri inaugurated a great hunting
party. Those who were willing to aid in dethroning the king were to
wear a bunch of grass in the hat as a sign. The king was seized and
imprisoned, and there hanged himself. His sons also killed
themselves and Eul-bul was then elevated to the perilous pinnacle of
royalty.
It was about the beginning of this century also that the Japanese,
during one of those spasmodic periods of seeming friendship asked
the king of Sil-la to send a noble maiden of Sil-la to be their queen.
The king complied and sent the daughter of one of his highest
officials, A-son-geup-ri.
 Chapter IX.
Rise of Yŭn.... rebellion against China.... siege of Keuk Fortress raised.... Ko-gu-
ryŭ surrenders to Yŭn.... Ko-gu-ryŭ disarmed.... Japanese attack Sil-la.... Păk-
je’s victory over Ko-gu-ryŭ.... moves her capital across the Han.... Păk-je
people in Sil-la.... Yŭn is punished.... Buddhism introduced into Ko-gu-ryŭ....
and into Păk-je.... amnesty between Ko-gu-ryŭ and Păk-je.... but Ko-gu-ryŭ
continues the war.... Păk-je in danger.... envoy to Japan.... Ch’ŭm-nye usurps
the throne of Păk-je.... and is killed.... Sil-la princes rescued.... Ko-gu-ryŭ and
Păk-je receive investiture from China.... China’s policy.... Nul-ji’s reign.... Ko-
gu-ryŭ and Păk-je transfer their allegience.... Yŭn extinct.... beginning of
triangular war.... diplomatic relations.... Ko-gu-ryŭ falls from grace.... first war
vessel.... diplomatic complications.... Păk-je humiliated.... her capital moved.

We have now come to the events which marked the rise of the
great Yŭn power in Liao-tung. They are so intimately connected with
the history of Ko-gu-ryŭ that we must give them in detail. For many
years there had been a Yŭn tribe in the north but up to the year 320
it had not come into prominence. It was a dependency of the Tsin
dynasty of China. Its chiefs were known by the general name Mo
Yong. In 320 Mo Yong-we was the acting chief of the tribe. He
conceived the ambitious design of overcoming China and founding a
new dynasty. The Emperor immediately despatched an army under
Gen. Ch’oe-bi to put down the incipient rebellion. Ko-gu-ryŭ and the
U-mun and Tan tribes were called upon to render assistance against
the rebels. All complied and soon the recalcitrant chieftain found
himself besieged in Keuk Fortress and was on the point of
surrendering at discretion when an event occurred which, fortunately
for him, broke up the combination and raised the siege. It was
customary before surrendering to send a present of food to the one
who receives the overtures of surrender. Mo Yong-we, in pursuance
of this custom, sent out the present, but for some reason it found its
way only into the camp of the U-mun forces while the others
received none. When this became known the forces of Ko-gu-ryŭ,
believing that Mo Yong-we had won over the U-mun people to his
side, retired in disgust and the Chinese forces, fearing perhaps a
hostile combination, likewise withdrew. The U-mun chiefs resented
this suspicion of treachery and vowed they would take Mo Yong-we
single-handed. But this they could not do, for the latter poured out
upon them with all his force and scattered them right and left. From
this point dates the rise of Yŭn. Gen. Ch’oe-bi fearing the wrath of
the Emperor fled to Ko-gu-ryŭ where he found asylum. Here the
affair rested for a time. The kingdom of Yŭn forebore to attack Ko-
gu-ryŭ and she in turn was busy strengthening her own position in
view of future contingencies. Ten years passed during which no
events of importance transpired. In 331 Eul-bul the king of Ko-gu-
ryŭ died and his son Soé began his reign by adopting an active
policy of defense. He heightened the walls of P’yŭng-yang and built
a strong fortress in the north, called Sin-sŭng. He followed this up by
strengthening his friendly relations with the court of China. These
facts did not escape the notice of the rising Yŭn power. Mo Yong-
whang, who had succeeded Mo Yong-we, hurled an expedition
against the new Sin-sŭng Fortress and wrested it from Ko-gu-ryŭ.
The king was compelled, much against his will, to go to Liao-tung
and swear fealty to the Yŭn power. Two years later the capital was
moved northward to Wan-do, in the vicinity of the Eui-ju of today.
This was done probably at the command of Yŭn who desired to have
the capital of Ko-gu-ryŭ within easy reach in case any complications
might arise.
Mo Yong-whang desired to invade China without delay but one of
his relatives, Mo Yong-han, advised him to disarm Ko-gu-ryŭ and the
U-mun tribe so that no possible enemy should be left in his rear
when he marched into China. It was decided to attack Ko-gu-ryŭ
from the north and west, but the latter route was to be the main
one, for Ko-gu-ryŭ would be expecting the attack from the north.
The strategem worked like a charm. Mo Yong-han and Mo Yong-p’ă
led a powerful army by way of the sea road while General Wang-u
led a decoy force by the northern route. The flower of the Ko-gu-ryŭ
army, 50,000 strong, marched northward under the king’s brother
Mu to meet an imaginary foe while the king with a few undisciplined
troops held the other approach. As may be supposed, the capital fell
speedily into the enemy’s hands but the king escaped. The Ko-gu-
ryŭ forces had been successful in the north and might return any
day, so the Yun forces were forbidden to go in pursuit of the king. To
insure the good behavior of the king, however, they burned the
palace, looted the treasure, exhumed the body of the king’s father
and took it, together with the queen and her mother, back to the
capital of Yŭn. With such hostages as these Yŭn was safe from that
quarter. The next year the king offered his humble apologies and
made a complete surrender, in view of which his father’s body and
his queen were returned to him but his mother-in-law was still held.
The same year Ko-gu-ryŭ moved her capital back to P’yŭng-yang. A
few years later by sending his son as substitute he got his mother-
in-law out of pawn.
In 344 new complications grew up between Sil-la and Japan. The
Japanese having already obtained one Sil-la maiden for a queen
made bold to ask for a royal princess to be sent to wed their king.
This was peremptorily refused and of course war was the result. A
Japanese force attacked the Sil-la coastguard but being driven back
they harried the island of P‘ung-do and finally worked around until
they were able to approach the capital. Finding the gates fast shut
they laid siege to the city. But their provisions were soon exhausted
and they were compelled to retire. Then the Sil-la forces swarmed
out and attacked them in the rear and put them to an ignominious
flight. Some years later the Japanese made a similar attempt but
were outwitted by the Sil-la soldiers who made manikins of grass to
represent soldiers, and the Japanese, seeing these, supposed that
Sil-la had been reinforced and so retired from the contest.
Ko-gu-ryŭ had been so severely handled by her northern neighbor
that she gave up for the time being her plans of conquest in that
direction. Instead of this she turned her attention toward her
southern neighbor Păk-je whose territory was a morsel not to be
despised. About the year 360 she erected a fort at Ch’i-yang not far
from the Păk-je capital which was then at Nam-han. Into this she
threw a large force consisting of 20,000 infantry and cavalry. They
began a systematic plundering of Păk-je. The army of the latter,
under the leadership of the Crown Prince, fell suddenly upon this fort
and gained a victory, for, when the Ko-gu-ryŭ forces retired, they left
5,000 dead upon the field. Păk-je followed up this victory by
throwing up a line of breastworks along the southern bank of the
Han river to insure against a future surprise on the part of her
unscrupulous northern neighbor. But Păk-je’s victories had shown her
the weakness of Ko-gu-ryŭ and reprisals were therefore in order. She
equipped an army of 30,000 men and penetrated the country of the
enemy. She met no resistance until her army stood beneath the
walls of P‘yŭng-yang. An attempt was made to storm the town,
during which the king of Ko-gu-ryŭ was mortally wounded by an
arrow, but the assault failed and the Păk-je army withdrew in good
order. The king of Păk-je, elated over so many evidences of his
growing power, promptly moved his capital across the Han River into
Ko-gu-ryŭ territory. Some say he settled at Puk-han the great
mountain fortress back of Seoul while others say he settled at Nam
P’yŭng-yang or “South P’yŭng-yang,” by which is meant the present
city of Seoul. Others still say it was at a point a short distance
outside the east gate of Seoul. But in spite of the apparent
successes of Păk-je it appears that the people were not satisfied. It
may be that military exactions had alienated their good will, or it
may be that they saw in these ambitious advances the sure presage
of speedy punishment at the hands of Ko-gu-ryŭ; but whatever the
cause may have been over a thousand people fled from Păk-je and
found asylum in Sil-la. The king set aside six villages as their place of
residence, and when Păk-je demanded to have them sent back
answer was returned that Sil-la could not drive from her borders
those who had sought asylum from the ill-treatment of Păk-je.
Three years before this, in 372, the Chinese had gained a signal
victory over the Yŭn kingdom and its king, Mo Yong-p’ung, had fled
for safety to Ko-gu-ryŭ. It must have been his last resource, for he
was likely to find little sympathy there. And so it proved for the king
immediately seized him and sent him a captive to China.
 The year 372 beheld an event of prime importance in the history
of Ko-gu-ryŭ and of the whole peninsula. It was the introduction of
Buddhism. It is probable that before this time some knowledge of
Buddhism was current in Korea, but as it is eminently a sacerdotal
institution but little more than indefinite reports could have been
circulated previous to the coming of the monks. We are not told
whether this was done at the request of Ko-gu-ryŭ or whether it was
at the advice of Pu-gyŭn, one of the petty kings who then divided
between them the north of China. Be that as it may, in 372 A.D.
images of Buddha were brought by a monk, Sun-do, and also a
Buddhist book called Pul-gyŭng. For this the king of Ko-gu-ryŭ
returned hearty thanks and forthwith set his son and heir to learning
the new doctrine. At the same time he gave an impetus to the study
of the Confucian code. It is quite probable that to this new departure
is due the fact that the next year the laws of the country were
overhauled and put in proper shape for use. In 375 two great
monasteries were built in the capital of Ko-gu-ryŭ. They were called
Cho-mun and I-bul-lan. It should be noticed that the introduction of
Buddhism into Korea was a government affair. There had been no
propagation of the tenets of this cult through emmisaries sent for
the purpose, there was no call for it from the people. In all
probability the king and his court were pleased at the idea of
introducing the stately ceremonial of the new faith. In fact it was a
social event rather than a religious one and from that date to this
there has not been a time when the people of Korea have entered
heartily into the spirit of Buddhism, nor have her most distinguished
representatives understood more than the mere forms and trappings
of that religion which among all pagan cults is the most mystical.
Păk-je was not long in following the example of her powerful
neighbor. In the year 384 a new king ascended the throne of Păk-je.
His name was Ch’im-yu. One of his first acts was to send an envoy
to China asking that a noted monk named Mararanta be sent to Păk-
je to introduce the Buddhist ritual. We notice that this request was
sent to the Emperor Hyo-mu (Hsia-wu), the proper head of the
Eastern Tsin dynasty, while Ko-gu-ryŭ had received hers at the
hands of one of those petty kings who hung upon the skirts of the
weakening dynasty and waited patiently for its dissolution. Each of
these petty states, as well as the central government of the Tsin,
was on the lookout for promising allies and such a request as this of
Păk-je could scarcely be refused. Mararanta, whose name smacks of
the south and who certainly cannot have been a Chinaman, was sent
to the Păk-je capital. He was received with open arms. His
apartments were in the palace where he soon erected a Buddhist
shrine. Ten more monks followed him and Buddhism was firmly
established in this second of the three Korean states. The greatest
deference was paid to these monks and they were addressed by the
honorific title To-seung. Sil-la received Buddhism some fifty years
later.
All this time fighting was almost continuous along the Ko-gu-ryŭ-
Păk-je border. The latter stood on the defensive and found it
necessary in 386 to build a line of breastworks along the border,
extending from Ch’ŭng-mok-yŭng northward to P’al-gon-sung and
thence westward to the sea. An amnesty was brought about through
a happy accident. A groom who had accidentally broken the leg of a
Păk-je prince’s horse had fled to Ko-gu-ryŭ to escape punishment.
Returning now to Păk-je, he purchased pardon by informing the king
that if, in battle, the Păk-je forces should direct their whole force
against that part of the enemy’s line where they should see a red
flag flying they would surely be successful. This turned out to be
true and Păk-je was once more successful, but followed up her
success only to the extent of securing a definite cessation of
hostilities and the erection of a boundary stone at Su-gok-sŭng to
witness forever against him who should dispute the point. But when
King Ch’im-yu of Ko-gu-ryŭ died in 392 and his son Tam-dok came
into power all previous obligations were swept away and he
proceeded to reopen the wound. He attacked Păk-je fiercely and
took ten of her towns. Then he turned northward and chastised the
Kŭ-ran tribe. When this was done he came back to the charge again
and seized Kwang-nu Fortress. This was an almost inaccessible
position on a high rock surrounded by the sea, but the hardy soldiers
of Ko-gu-ryŭ after twenty days of siege found seven paths by which
the wall could be reached, and they finally took the place by a
simultaneous assault at these various points. When the court of Păk-
je heard of this well-nigh impossible feat, all hope of victory in the
field was taken away, and they could only bar the gates of the
capital and await the turn of events. This king, Tam-dok, was as
enthusiastically Buddhistic as his father. He made a decree that all
the people of Ko-gu-ryŭ should adopt the Buddhistic faith and a few
years later built nine more monasteries in P’yŭng-yang.
A year later King A-sin of Păk-je sent his son, Chön-ji, to Japan as
an envoy. It is likely, but not certain, that it was a last resource of
Păk-je to secure help against Ko-gu-ryŭ. This is the more likely from
the fact that he went not only as an envoy but also as a hostage, or
a guarantee of good faith. If this was the hope of Păk-je it failed, for
no Japanese army was forthcoming. As another means of self-
preservation King A-sin formed a great school of archery, but the
people did not like it; for exercise in it was compulsory, and many of
the people ran away.
In 399 Ko-gu-ryŭ sent an envoy to the Yŭn capital to pay her
respects, but the king of that country charged Ko-gu-ryŭ with
ambitious designs and sent an army of 30,000 men to seize the
fortresses of Sin-sŭng and Nam-so, thus delimiting the frontier of
Ko-gu-ryŭ to the extent of 700 li. They carried back with them 5,000
“houses,” which means approximately 25,000 people, as captives. It
is difficult to believe this enumeration unless we conclude that it
means that the people living within the limit of the 700 li were taken
to be citizens of Yŭn.
The fifth century of our era dawned upon a troubled Korea. The
tension between the three rival powers was severe, and every nerve
was strained in the struggle for preeminence. In 402 Nă-mul, the
king of Sil-la, died and Sil-sŭng came to the throne. He sent out
feelers in two directions, one toward Ko-gu-ryŭ in the shape of a
hostage, called by euphemism an envoy, and another of the same
sort to Japan; which would indicate that Sil-la was still suffering from
the depredations of the Japanese corsairs. The envoy to Ko-gu-ryŭ
was the king’s brother, Pok-ho, and the one to Japan was also his
brother, Mi-sa-heun. We remember that Păk-je already had an envoy
in Japan in the person of the king’s eldest son Chön-ji. Now in 405
the king of Păk-je died. Chön-ji was the rightful heir but as he was in
Japan the second son should have assumed the reins of
government. As a fact the third son Chŭng-nye killed his brother and
seized the scepter. Hearing of his father’s death, Chön-ji returned
from Japan with an escort of a hundred Japanese, but learning of his
brother’s murder he feared treachery against himself and so landed
on an island off the coast where he remained until the people, with
a fine sense of justice, drove Ch’ăm-nye from the throne and
welcomed back the rightful heir.
Meanwhile interesting events were transpiring in Sil-la. In 403 Sil-
sung, King of that land, fearing lest harm overtake his two brothers
whom he had sent the year before to Ko-gu-ryŭ and Japan, was
seeking for some means of getting them back. This might not be an
easy thing to do, for to ask their return so soon would perhaps
arouse the suspicion of these neighbors, and precipitate a war. Ko-
gu-ryŭ had often taken up arms for a less affront than this. An
official, Pak Che-san, volunteered to undertake this delicate mission
even though it cost him his life. He went first to Ko-gu-ryŭ and there
proved so skillful a diplomat that he soon brought Prince Pok-ho
back to Sil-la. The mission to Japan was a different matter, but he
was equal to the occasion. Before starting out he said to the king: “I
will bring the Prince back though it cost my life; only, before I go, I
must ask you to imprison my family; otherwise I cannot succeed.”
The king acceded to this strange request and Pak Che-san, starting
immediately as if in flight, without even changing his garments, fled
until he came to the Yul Harbor. Even his wife he repulsed,
exclaiming “I have determined to die.” He apparently feared that the
sight of her might shake his loyal purpose. He arrived in Japan as a
political fugitive, but the king suspected him until news came that
his family had been imprisoned. This seemed to prove his statement
and he was received graciously. He pretended that he wished to lead
a Japanese force against Sil-la. Mi-sa-heun, the Prince whom he had
come to rescue, was in the secret and heartily seconded the plan.
The king made them joint leaders of an expedition. The fleet arrived
at a certain island and there Pak succeeded in spiriting Mi-sa-heun
away by night in a little boat while he himself remained behind, to
delay the inevitable pursuit. Mi-sa-heun begged him with tears to
accompany him but he refused to jeopardise Mi-sa-heun’s chances of
escape by so doing. In the morning he pretended to sleep very late
and no one suspected the flight of the Prince until late in the day
when concealment was no longer possible. When the Japanese
found that they had been duped they were in a terrible rage. They
bound Pak and went in pursuit of the run-away. But a heavy fog
settled upon the sea and frustrated their plan. Then they tortured
their remaining victim and to their inquiries he replied that he was a
loyal subject of Kye-rim (the name of Sil-la at that time) and that he
would rather be a Kye-rim pig than a subject of Japan; that he
would rather be whipped like a school-boy in Kye-rim than receive
office in Japan. By these taunts he escaped a lingering death by
torture. They burned him alive there on the island of Mok-do. When
the king of Sil-la heard of his brave end he mourned for him and
heaped upon him posthumous honors, and Mi-sa-heun married his
preserver’s daughter. The wife of the devoted Pak ascended the pass
of Ap-sul-yŭng whence she could obtain a distant view of the islands
of Japan. There she gave herself up to grief until death put an end
to her misery.
In 413 a new king came to the throne of Ko-gu-ryŭ. called Kö-
ryŭn. As China and Ko-gu-ryŭ had been kept apart by the
intervening Yun, and had acquired some power of sympathy through
mutual fear of that power, we are not surprised that the new king of
Ko-gu-ryŭ condescended to receive investiture from the Emperor, nor
that the latter condescended in turn to grant it. It was formally
done, and the act of Ko-gu-ryŭ proclaimed her vassalage to China.
From that time on excepting when war existed between them, the
kings of Ko-gu-ryŭ were invested by the Emperor with the insignia of
royalty. Two years later the Emperor conferred the same honor upon
the king of Păk-je. It was always China’s policy to keep the kingdoms
at peace with each other so long as they all wore the yoke of
vassalage; but so soon as one or the other cast it off it was her
policy to keep them at war.
In 417 Nul-ji came to the throne of Sil-la and began a reign that
was to last well on toward half a century. He was a regicide. He had
been treated very harshly by the king and had more than once
narrowly escaped with his life. It is therefore the less surprising,
though none the less reprehensible, that when the opportunity
presented of paying off old scores he succumbed to the temptation.
He ascended the throne not with the title of I-sa-geum, which had
been the royal title for centuries, but with the new title of Ma-rip-
kan. However doubtful may have been his title to the crown his
reign was a strong one. Among the far-reaching effects of his reign
the introduction of carts to be drawn by oxen was the most
important.
The friendly relations of Ko-gu-ryŭ with the Tsin dynasty were cut
short by the extinction of that dynasty in 419 but in 435 Ko-gu-ryŭ
made friendly advances toward the Northern Wei dynasty and,
finding sufficient encouragement, she transferred her allegience to
that power. Meantime Păk-je had transferred hers to the Sung
dynasty which arose in 420.
It was in 436 that P’ung-hong, the “Emperor” of Yun, found
himself so weak that he could not withstand the pressure from the
Chinese side and asked the king of Ko-gu-ryŭ to grant him asylum.
Consent was given and an escort was sent to conduct him to the Ko-
gu-ryŭ capital. He found that this sort of life had its drawbacks; for,
to begin with, the king did not address him as emperor but simply as
king. This was a great affront to his dignity and, though he was
treated very handsomely, he assumed such a supercillious bearing
that the king had to curtail his retinue and his income. He had been
given quarters in Puk-p’ung and from there the mendicant emperor
applied to the Sung Emperor for asylum. It was granted, and seven
thousand soldiers came to escort him; but ere they arrived the king
of Ko-gu-ryŭ sent two generals, Son-su and Ko-gu, who killed the
imperial refugee and nine of his attendants. The Sung troops,
arriving on the instant, discovered the crime and caught and
executed the two generals who had perpetrated it.
In 449 a Ko-gu-ryŭ general was out on a hunting expedition and
the chase brought him into Sil-la territory near the present town of
Kang-neung. The prefect of the district, in an excess of patriotic
enthusiasm, seized him and put him to death. An envoy came in
haste to the Sil-la capital demanding why this outrage had been
committed. War would have been declared on the spot had not Sil-la
been profuse in apologies. She might have spared herself this
humiliation for war was sure to break out soon in any case. When ng
came to the throne of Păk-je in 455, Ko-gu-ryŭ took advantage of
the confusion, consequent upon the change, to attack her. Sil-la,
who, though ordinarily a peaceful power, had been perforce drawn
into war-like operations and had acquired some military skill, now
sided with Păk-je. Sending a considerable number of troops she
reinforced Păk-je to the extent of warding off the threatened
invasion. But Păk-je, though glad to find herself extricated from her
position of danger, would allow no feelings of gratitude to stand in
the way of her ancient feud against Sil-la; so this act of friendship
not only did not help toward peace but on the contrary, by showing
Sil-la the fickleness of Păk-je, made peace all the more impossible.
The middle of the fifth century marks the point when all friendly
relations between the three Korean states were broken off and an
actual state of war existed between them from this time on, though
active military operations were not constant. This we may call the
Triangular War.
The key to this great struggle, which resulted in the advancement
of Sil-la to the control of the whole peninsula, lay not so much in the
relative military strength of the three rival kingdoms as in the skill
which each developed in diplomacy. Each was trying to gain the
active support of China, knowing very well that if China should once
become thoroughly interested in favor of any one of the three
powers the other two would be doomed.
We will remember that Ko-gu-ryŭ had cultivated friendly relations
with the Sung dynasty while Păk-je had made herself agreeable to
the Wei dynasty. In this Păk-je chose the wiser part for the Wei
power was nearer and more powerful. In 466 Ko-gu-ryŭ lost a
splendid opportunity to establish herself in the good graces of the
Wei Emperor, and so insure her preeminence in the peninsula. The
Emperor Hsien-wen made friendly advances and requested the
daughter of the king of Ko-gu-ryŭ for his wife. With a short-
sightedness that is quite inexplicable this request was put off by the
lame excuse that his daughter was dead. This being easily proved a
falsehood, Ko-gu-ryŭ fell from the good graces of the very power
whose friendship she should have cultivated.
The year 467 witnessed an important innovation in Korea. Sil-la
took the lead in the construction of war vessels. The one made at
that time was doubtless intended for use against the Japanese
corsairs. That Sil-la had been gaining along military lines is shown by
her successful repulse of a Ko-gu-ryŭ invasion in this year, in which
the wild people of some of the Mal-gal tribes assisted Ko-gu-ryŭ.
After the latter had been driven back, Sil-la built a fortress at Po-eun
on her northern border to guard against a repetition of this invasion.
Ko-gu-ryŭ and Păk-je were now exerting themselves to the utmost
to make capital out of their Chinese alliances. Ko-gu-ryŭ sent rich
presents and richer words to the Sung capital and so won the
confidence of that power. Păk-je, on the other hand, sent word to
the Wei Emperor that Ko-gu-ryŭ was coquetting with the Sung court
and with the wild Mal-gal tribes, insinuating that this was all
detrimental to the interests of Păk-je’s patron.
As this was without result, she sent and asked openly that the Wei
Emperor send an army and chastise Ko-gu-ryŭ. The Emperor replied
that until Ko-gu-ryŭ committed some overt act of more hostile
import than the mere cementing of peaceful alliances no notice
could be taken of her. In other words the Wei power refused to be
the aggressor, much to Păk-je’s chagrin. The Wei Emperor sent this
answer by way of Ko-gu-ryŭ and the king of that country was
ordered to grant the messenger a safe conduct through his territory.
But Ko-gu-ryŭ, as though bent on self-destruction, refused to let him
pass, and so the great northern kingdom approached one step
nearer the precipice which was to prove her destruction. Upon
learning the news of this affront the Emperor was highly incensed
and tried to send the messenger by way of a southern port; but
stress of weather rendered this impossible and Păk-je, receiving no
answer to her missive, took offense and would have nothing more to
do with China, for a time. By the time she had recovered her temper,
Ko-gu-ryŭ had in some way patched up her difficulty with the Wei
court and so scored a point against Păk-je. And for a time she was
on friendly terms with both the Wei and Sung dynasties.
At this point Ko-gu-ryŭ decided upon a bold attempt to swallow
Păk-je bodily. It was to be done partly by strategem and partly by
force. A monk of Ko-gu-ryŭ named To-rim, a fellow of excellent craft,
arrived at the Păk-je capital as if seeking refuge. The king received
him with open arms and, finding him an excellent chess player, made
him his trusty councilor. This monk told the king that the palaces,
walls, tombs and public buildings ought to be thoroughly repaired,
and so induced him to drain the public treasury in this work, and
also in bringing a huge monolith from Uk-nyi to the capital. This
done the monk fled back to Ko-gu-ryŭ and announced that the
treasury of Păk-je was empty and it was a good time to attack her. A
large army was put in the field, guided by one Kŭl-lu, a Păk-je
fugitive from justice. Almost before Păk-je was aware, her capital
was surrounded. She had applied to Sil-la for help, but too late. First
the suburbs were laid in ashes, and then access being gained, the
palace was fired. The king fled with ten attendants out the west
gate, but Kŭl-lu the renegade followed and overtook him. The king
begged for mercy upon his knees but Kŭl-lu spit thrice in his face,
bound him and sent him to the fortress of A-han where lie was
killed. Then the Ko-gu-ryŭ army went back north carrying with them
8,000 captives, men and women.
Meanwhile Prince Mun-ju had obtained help from Sil-la and with
10,000 troops was hastening homewards. He found the city in
ashes, his father dead, the people mourning their lost, who had
been dragged away captive. He promptly assumed control of affairs,
moved the capital southward to Ung-jin the present Kong-ju, took all
the Păk-je people away from Han-yang (Seoul) and moved them
back across the Han River and abandoned all the territory beyond
that natural barrier to Ko-gu-ryŭ to whom it had originally belonged.
The following year he tried to send a message to the Sung Emperor
by way of Ko-gu-ryŭ but the messenger was intercepted and the
message stopped.
 Chapter X.
Quelpart.... origin of T’am-na.... new alliances.... advances in Sil-la.... but not in
Păk-je nor Ko-gu-ryŭ.... temporary peace.... Buddhism in Sil-la.... remnants of
barbarism.... influence of Chinese literature.... important reforms.... Ko-gu-
ryŭ’s foreign relations.... conquest of Dagelet Island.... posthumous titles....
colors in official grades.... Wei displeased.... the “miracle” of Yi Cha-don....
end of Ka-rak.... Sil-la rejects Chinese calendar.... confusion in China.... Păk-je
attempts reform.... history of Sil-la.... two alliances.... Păk-je and Ko-gu-ryŭ
envoys to China.... advance of Buddhism in Sil-la.... music in Sil-la.... war
between Păk-je and Sil-la.... retrogression in Sil-la because of Buddhism....
Ko-gu-ryŭ and the Sui Emperor.... the Ondali.

Tradition says that in the dawn of history when the island of Che-
ju (Quelpart) was covered only with a tangled forest three sages
arose from a crevice in the ground. This spot is shown to this day by
the people of Che-ju. These three men were Ko-ŭlla, Yang-ŭlla and
Pu-ŭlla. As they stood upon the shore they saw three stout chests
floating in from the south-east. Drawing them to land and opening
them the three wise men discovered that each chest contained a
calf, a colt, a dog, a pig and a woman, together with sundry seeds,
such as beans, wheat, barley, millet and rice. By the three families
thus organised the island was populated. During the early days of
Sil-la a certain court astrologer announced that the “Friend Star” was
visible in the south and that a distinguished visitor would soon
arrive. Soon after this three men came by boat from Quelpart,
landing at the harbor of T’am-jin, now Kang-jin. They came straight
to the court of Sil-la where they were hospitably entertained. One of
the visitors was Ko-hu, one was Ko-ch’ŭng, but the name of the third
is lost. The king called the first Sŭng-ju or “Lord of the Star,” the
second Wang-ja or “King’s Son” and the third To-nă or “The One
who has Come.” He named their country T’am from the name of the
port where they landed, and na, which seems to have meant
“Kingdom”, for we find that the last syllable of Sil-la is this same na
changed by euphonic laws to la. It is the root of the present Korean
word na-ra or “kingdom.” So the kingdom was called T’am-na. The
authorities are at a loss to tell the date or even the reign during
which these events transpired. In the year 477 the little kingdom of
T’am-na sent an envoy to the court of Păk-je with gifts. This is the
first really authentic mention of the place. If tradition is of any value
it must be confessed that the story of the peopling of Quelpart
points toward a southern origin.
In 479 the aged king of Ko-gu-ryŭ, Kö-ryŭn, now in the sixty-
eighth year of his reign, sought and obtained recognition from
Emperor Ko-je (Kao-ti) the founder of the Ch’i dynasty in China. That
this occurred in the very first year after the founding of that dynasty
shows how sedulously Ko-gu-ryŭ was cultivating the good-will of the
Chinese. Păk-je was not far behind, for she swore allegiance to the
same Emperor only two years later.
During all these years it is to Sil-la that we must look for any signs
of internal improvement, any of those innovations which are the
mile-stones of progress. We saw above how she introduced the use
of the cart and so raised a great burden from the shoulders of the
people. The wheel is the great burden bearer of history. And now we
find her introducing further reforms. The first was the horse relay
system called the yong-ma. It did not bear so directly upon the
condition of the people but it afforded an opportunity for the rapid
transmission of official information and thus indirectly had an
important bearing upon the welfare of the masses. In the next place,
she organised a general market where at stated intervals merchants
from the various districts could meet and exchange commodities.
These are things that we look upon as matters of course and we do
not realise their importance till we imagine ourselves deprived of the
comforts that spring from the possibility of rapid communication and
exchange of commodities. That Ko-gu-ryŭ had not made similar
advances in the line of industrial reform is shown by the fact that
when the Emperor of the Wei dynasty sent to grant investiture to
Na-un the twenty-first king of Ko-gu-ryŭ in 499 he presented him
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookultra.com