DISTRIBUTED SYSTEMS

Introduction: Part 3
(Service Models)

1
Service Models (Application Architectures)

2
Centralized model
• No networking
• Traditional time-sharing system
• Single workstation/PC or direct connection of multiple terminals to a
computer
• One or several CPUs
• Not easily scalable
• Limiting factor: number of CPUs in system
– Contention for same resources (memory, network, devices)

3
Client-Server model
• Clients send requests to servers
• A server is a system that runs a service
• Clients do not communicate with other clients

4
Layered architectures in software
design
•Break functionality into multiple layers
• Each layer handles a specific abstraction
– Hides implementation details and specifics of hardware, OS, network
abstractions, data encoding, …

Applications
Includes naming, security, persistence, notifications, agreement,
Middleware remote procedures, data encoding, …

Includes layering for

Operating System file systems, networking, devices, memory

Hardware

5
Tiered architectures in networked
systems
•Tiered (multi-tier) architectures
– Distributed systems analogy to a layered architecture
• Each tier (layer)
– Runs as a network service
– Is accessed by surrounding tiers

The basic client-server architecture is a two-tier

model

6
Multi-tier example

middle
client back end
tier

User interface • Queuing requests • Database system

Data • Coordinating a • Legacy software
presentation transaction among
& validation multiple servers
• Managing
connections
• Formatting/
converting data

7
Multi-tier example

object
application store
client web server
server
database

8
Multi-tier example

Some tiers may be transparent to the application

object

load balancer
store

web server

application
server
firewall

firewall

cache
client

database

9
Microservices Model
Data storage
Collection of autonomous
service
services
Sensor
Data Logging
normalization
service
service • Each service:
Sensor Data storage – Runs independently
service
– Has a well-defined
interface
Caching
service
Data analytics
service
– May be shared by
multiple applications
Client access
service
Web client
service
• Main application
coordinates interactions

10
Peer-to-Peer (P2P)
Model
•No reliance on servers
• Machines (peers) communicate with each other
• Goals
server
– Robustness
– Self-scalability clients

• Examples
– BitTorrent, Skype

peers
11
Hybrid model
• Many peer-to-peer architectures still rely on a server
– Look up, track users
– Track content
– Coordinate access
• But traffic-intensive workloads are delegated to peers

Images from: http://clipart-library.com/laptop-cliparts.html

12
Processor pool model
• Collection of CPUs that can be assigned processes on demand
• Similar to hybrid model
– Coordinator dispatches work requests to available processors
• Render farms, big data processing, machine learning

13
Cloud Computing
Resources are provided as a network (Internet)
service
Software as a Service (SaaS)
Remotely hosted software: email, productivity, games, …
Salesforce.com, Google Apps, Microsoft 365

Platform as a Service (PaaS)

Execution runtimes, databases, web servers, development environments, …
Google App Engine, AWS Elastic Beanstalk

Infrastructure as a Service (IaaS)

Compute + storage + networking: VMs, storage servers, load balancers
Microsoft Azure, Google Compute Engine, Amazon Web Services

Storage
Remote file storage
• Dropbox, Box, Google Drive, OneDrive, … 14
The End

15