Module-3

Cloud Platform Architecture

over Virtualized Data Centers
Reference: Distributed and Cloud Computing From Parallel
Processing to the Internet of Things, Kai Hwang Geoffrey C.
Fox, and Jack J. Dongarra, Morgan Kaufmann Oc 2012
Elsevier, Inc. All rights reserved.

1
 Cloud Computing and Service Models
• This section deals with the cloud platform
architecture, service models, and programming
environments.
• Users can access and deploy cloud applications
from anywhere in the world at very competitive
costs.
• Virtualized cloud platforms are often built on top
of large data centers.
’ Clouds aim to power the next generation of data
centers by architecting them as virtual resources over
automated hardware, databases, user interfaces, and
application environments.
 Public, Private, and Hybrid Clouds
• The concept of cloud computing has evolved from cluster,
grid, and utility computing:
—Cluster and grid computing leverage the use of
many computers in parallel to solve
problems of any size.
—Utility and Software as a Service (SaaS) provide computing
resources as a service with the notion of
pay per use.
—Cloud computing leverages dynamic resources to deliver
large numbers of services to end users.
—Cloud computing is a high-throughput computing (HTC)
paradigm whereby the infrastructure provides the services
through a large data center or server farms.
—The cloud computing model enables users to share access to
resources from anywhere at any time through their
connected devices.
 Public, Private, and Hybrid Clouds
• The cloud will free users to focus on user
application development by outsourcing job
execution to cloud providers:
—In this scenario, the computations (programs) are sent to
where the data is located, rather than copying the
data to millions of desktops.
—Cloud computing avoids large data movement, resulting in
much better network bandwidth utilization.
—Furthermore, machine virtualization has enhanced
resource utilization, increased application
flexibility, and reduced the total cost of using virtualized
data-center resources.
 Centralized versus Distributed
Computing
• All computations in cloud applications are distributed
to servers in a data center.
— These are mainly virtual machines (VMs) in virtual clusters
created out of data-center resources.
• Cloud platforms are systems distributed through
virtualization. As Figure 4.1 shows, both public clouds
and private clouds are developed in the Internet:
—Commercial cloud providers such as Amazon,
Google,
and Microsoft created their platforms to
be distributed geographically.
—This distribution is partially attributed to
fault
tolerance, response latency reduction, and even legal
reasons.
Centralized versus Distributed
Computing
 Public
• Clouds
A public cloud is built over the Internet and can be
accessed by any user who has paid for the
service.
—Public clouds are owned by service providers and are
accessible through a subscription.
• The callout box in top of Figure 4.1 shows the
architecture of a typical public cloud.
—Many public clouds are available, including Google App
Engine (GAE), Amazon Web Services (AWS),
Microsoft Azure, IBM Blue Cloud, and
Salesforce.com’s Force.com.
—The providers of the aforementioned clouds are
commercial providers that offer a publicly accessible
remote interface for creating and managing VM
instances .
 Private Clouds
• A private cloud is built within the domain of
an intranet owned by a single organization.
• Therefore, it is client owned and managed, and its
access is limited to the owning clients and their
partners.
— Its deployment was not meant to sell capacity over the
Internet through publicly accessible interfaces.
• A private cloud is supposed to deliver more efficient
and convenient cloud services.
• It may impact the cloud standardization, while
retaining greater customization and organizational
control.
 Hybrid
• A hybrid cloud isClouds
built with both public and private
clouds, as shown at the lower-left corner of Figure 4.1.
• Private clouds can also support a hybrid cloud model by
supplementing local infrastructure with computing
capacity from an external public cloud.
— For example, the Research Compute Cloud (RC2) is a private
cloud, built by IBM, that interconnects the computing and IT
resources at eight IBM Research Centers scattered throughout
the United States, Europe, and Asia.
• A hybrid cloud provides access to clients, the partner
network, and third parties.
 Data-Center Networking Structure
• The core of a cloud is the server cluster (or
VM cluster).
• The gateway nodes provide the access points of the
service from the outside world.
—These gateway nodes can be also used for security
control of the entire cloud platform.
• In physical clusters and traditional grids, users
expect static demand of resources:
—Clouds are designed to handle fluctuating workloads, and
thus demand variable resources dynamically.
• Private clouds will satisfy this demand if properly
designed and managed.
 Data-Center Networking Structure
° Data centers and supercomputers also differ in
networking requirements, as illustrated in Figure 4.2.
° Supercomputers use custom-designed high-bandwidth
networks such as fat trees or 3D torus networks.
—Data center networks are mostly IP-based commodity
networks, such as the 10 Gbps Ethernet network, which is
optimized for Internet access.
° Figure 4.2 shows a multilayer structure for accessing the
Internet:
— The server racks are at the bottom Layer 2, and
they are connected through fast switches (S) as the
hardware core.
— The data center is connected to the Internet at Layer
3 with many access routers (ARs) and border routers (BRs).
Data-Center Networking Structure
 Cloud Development Trends
• Although most clouds built since 2010 are
large public clouds, the authors believe private
clouds will grow much faster than public clouds
in the future:
—Private clouds are easier to secure and more
trustworthy within a company or
organization.
—Once private clouds become mature and better secured,
they could be open or converted to public clouds.
—Therefore, the boundary between public and private
clouds could be blurred in the future.
• Most likely, most future clouds will be hybrid in nature.
13
 Cloud Development Trends
— For example, an e-mail application can run in the
service-access nodes and provide the user interface for
outside users; the application can get the service from
the internal cloud computing services (e.g., the e-mail
storage service).
• These nodes are called runtime supporting
service nodes.
— For example, there might be distributed locking
services for supporting specific applications.
• Finally, it is possible that there will be some
independent service nodes:
— Those nodes would provide independent
services for other nodes in the cluster.
 Cloud Ecosystem and Enabling
Technologies
• Cloud computing platforms differ from
conventional computing platforms in many
aspects:
— The traditional computing model involves
buying the HW, acquiring the necessary system SW,
installing the system, testing the configuration, and
executing the application code and management of
resources.
—What is even worse is that this cycle repeats itself in
about every 18 months, meaning the machine
we bought becomes obsolete every 18 months.
 Cloud Design Objectives
• The following list highlights six design objectives for cloud
computing:
- Shifting computing from desktops to data centers over the
internet.
—Service provisioning and cloud economics by providers
supply cloud services signing SLAs with consumers
and end users.
— Scalability in performance by improving software and
infrastructure services help to scale in performance.
—Data privacy protection by providing cloud as a
trusted services.
- High quality of cloud services by improving the QoS of cloud
to make clouds interoperable among multiple providers.
— New standards and interfaces by providing universally accepted
APIs and access protocols for high portability and flexibility
of virtualized applications.
 Cloud Ecosystems
• With the emergence of various Internet clouds, an
ecosystem of providers, users, and technologies has been
appeared.
• This ecosystem has evolved around public clouds:
—Strong interest is growing in open source cloud computing
tools that let organizations build their own infrastructure-as-a-
service (laas) clouds using their internal infrastructures.
• Private and hybrid clouds are not exclusive, since
public clouds are involved in both cloud types:
—A private/hybrid cloud allows remote access to its
resources
over the Internet using remote web service interfaces such as
that used in Amazon EC2.
—An ecosystem was suggested by Sotomayor (Figure 4.4) for
building private clouds.
 Cloud Ecosystems
• The Figure 4.4 suggested four levels of
ecosystem development in a private cloud:
—At the user end, consumers demand a flexible
platform.
—At the cloud management level, the cloud manager
provides virtualized resources over an IaaS platform.
— At the virtual infrastructure (VI) management level, the
manager allocates VMs over multiple server clusters.
— At the VM management level, the VM managers handle
VMs installed on individual host machines.
• An ecosystem of cloud tools attempts to span both
cloud management and VI management.
18
Cloud Ecosystems
 The Three Cloud Service Models
• Cloud computing delivers infrastructure, platform,
and software (application) as services, which are
made available as subscription-based services in
a pay-as- you-go model to consumers.
• The services provided over the cloud can be generally
categorized into three different service models:
—namely IaaS, Platform as a Service (PaaS), and
Software as a Service (SaaS).
— These form the three pillars on top of which
cloud computing solutions are delivered to end
users.
— All three models allow users to access services
over the Internet, relying entirely on the infrastructures
of cloud service providers.
20
 The Three Cloud Service
• Models
These models are offered based on various service —
level- agreements (SLAs) between providers and users:
— The SLA for cloud computing is addressed in terms of
service availability, performance, and data protection and
security.
— Figure 4.5 illustrates three cloud models at different service
levels of the cloud.
• Software as a service (SaaS) model is applied at the
application end using special interfaces by users or clients.
• At the platform as a service (Paa5) layer, the
cloud platform must perform billing services and handle
job queuing, launching, and monitoring services.
• At the bottom layer , the infrastructure as a
service (IaaS) services, databases, compute instances, the
file system, and storage .
The Three Cloud Service
Model5
 Infrastructure as a Service(IaaS)
• The IaaS model allows users to use virtualized IT
resources for computing, storage, and
networking:
—The user can deploy and run his applications over his
chosen OS environment.
• The user does not manage or control the underlying cloud
infrastructure, but has control over the OS, storage,
deployed applications, and possibly select networking
components.
—This IaaS model encompasses storage as a service,
compute instances as a service, and communication as
a service.
• The Virtual Private Cloud (VPC) in Figure 4.6 shows
how to provide Amazon EC2 clusters and S3 storage to
multiple users.
Infrastructure as a Service
 Platform as a Service (PaaS)
• PaaS cloud model allows to develop, deploy, and
manage the execution of applications using
provisioned resources demands a cloud
platform with the proper software
environment:
—Such a platform includes OS and runtime library
support.
—This has triggered the creation of the PaaS model to
enable users to develop and deploy their user
applications.
° Table 4.2 highlights cloud platform services
offered by five PaaS services.
25
Platform as a Service (PaaS)
 Platform as a Service (PaaS)
• The PaaS cloud model is an integrated computer
system consisting of both HW and SW infrastructure:
—The user application can be developed on this virtualized
cloud platform using some programming languages and
software tools supported by the provider (e.g., Java,
Python,
.NET).
— The user does not manage the
underlying cloud infrastructure.
— The cloud provider supports user application
development and testing on a well-defined service
platform.
° The PaaS model enables a collaborated software
development platform , SW management, integration,
se vice onitorin solutions fo use s f o
dif erent parts of the world.
Google App Engine for PaaS Applications
 Software as a Service
• SaaS(SaaS)
refers to browser-initiated application
software over thousands of cloud customers:
—The SaaS model provides SW applications as a service.
—As a result, on the customer side, there is no upfront
investment in servers or SW licensing.
—On the provider side, costs are kept rather low,
compared with conventional hosting of user applications.
—Customer data is stored in the cloud that is either vendor
proprietary or publicly hosted to support PaaS and
IaaS.
 Software as a Service
— (SaaS)
The best examples of SaaS services include
Google Gmail and docs, Microsofi SharePoint, and
the CRM software from Salesforce.com.
—Providers such as Google and Microsofi
offer integrated IaaS and PaaS
services.
— Whereas others such as Amazon and GoGrid
offer pure IaaS services and expect third-party
PaaS providers such as Manjrasofi to offer
application development and deployment
services on top of their infrastructure services.
 Architectural Design of Compute and
Storage Clouds
• An Internet cloud is envisioned as a public cluster of
servers provisioned on demand to perform collective
web services or distributed applications using data-
center resources.
• Scalability, virtualization, efficiency, and reliability are
four major design goals of a cloud computing platform:
—Cloud management receives the user request, finds the
correct resources, and then calls the provisioning
services which invoke the resources in the cloud.
— The cloud management software needs to
support both physical and virtual machines
(VMs).
— Security in shared resources and shared access of data
centers also pose another design challenge.
 Architectural Design of Compute and
Storage Clouds
• The platform needs to establish a very large-scale
HPC infrastructure.
—The HW and SW systems are combined to make it easy
and efficient to operate.
• System scalability can benefit from cluster
architecture: If one service takes a lot of processing
power, storage capacity, or network traffic, it is
simple to add more servers and bandwidth, or
data can be put into multiple locations.
—For example, user e-mail can be put in three disks which
expand to different geographically separate data
centers.
 Architectural Design of Compute and
Storage Clouds
• The key driving forces behind cloud computing
are
the ubiquity of broadband and wireless
networking, and progressive improvements in
Internet computing SW:
—Cloud users are able to demand more capacity at peak
demand, reduce costs, experiment with new
services,
and remove unneeded capacity, whereas service
providers can increase system utilization via
multiplexing, virtualization, and dynamic resource
provisioning.
—Clouds are enabled by the progress in HW, SW, and
networking technologies summarized in Table
4.3.
Architectural Design of Compute
and Storage Clouds
 A Generic Cloud Architecture
• Figure 4.14 shows a security-aware cloud architecture.
The Internet cloud is envisioned as a massive
cluster of servers:
— These servers are provisioned on demand to
perform collective web services or distributed
applications using data-center resources.
• The cloud platform is formed dynamically by
provisioning or deprovisioning servers, software, and
database resources.
—Servers in the cloud can be physical machines or VMs.
—User interfaces are applied to request services.
— The provisioning tool carves out the cloud system to
deliver the requested service.
A Generic Cloud Architecture
 A Generic Cloud Architecture
• The cloud computing resources are built into the data
centers, which are typically owned and operated by a
third-party provider:
• The cloud demands a high degree of trust of massive
amounts of data retrieved from large data centers.
• To build a framework to process large-scale data stored in
the storage system:
— This demands a distributed file system over the
database system.
— Other cloud resources are added into a cloud
platform, including storage area networks (SANs),
database systems, firewalls, and security devices.
 A Generic Cloud Architecture
• Web service providers offer special APIs that
enable developers to exploit Internet
clouds.
• The SW infrastructure of a cloud platform must
handle all resource management and do most
of the maintenance automatically:
—SW must detect the status of each node server joining
and leaving, and perform relevant tasks
accordingly.
—Cloud computing providers, such as Google and
Microsoft, have built a large number of data centers all
over the world.
—Each data center may have thousands of servers.
 A Generic Cloud Architecture
• In general, private clouds are easier to manage,
and public clouds are easier to access.
• The trends in cloud development are that more
and more clouds will be hybrid.
— This is because many cloud applications must go beyond
the boundary of an intranet.
• One must learn how to create a private cloud and
how to interact with public clouds in the open
Internet.
• Security becomes a critical issue in safeguarding the
operation of all cloud types.
39
 Layered Cloud Architectural
Development
• The architecture of a cloud is developed at three layers:
infrastructure, platform, and application as
demonstrated in Figure 4.15).
• The services to public, private, and hybrid clouds are
conveyed to users through networking support over
the Internet and intranets involved.
° It is clear that the infrastructure layer is deployed first to
support IaaS services.
° This infrastructure layer serves as the foundation for
building the platform layer of the cloud for supporting
PaaS services.
• The platform layer is a foundation for implementing the
application layer for SaaS applications.
Layered Cloud Architectural
Development
 Layered Cloud Architectural
Development
• The infrastructure layer is built with
virtualized compute, storage, and network
resources.
—Internally, virtualization realizes automated provisioning
of resources and optimizes the infrastructure
management process.
• The platform layer is for general-purpose and
repeated usage of the collection of SW
resources:
—This layer provides users with an environment to develop
their applications, to test operation flows, and to
monitor execution results and performance.
—The virtualized cloud platform serves as a “system
middleware” between the infrastructure and application
layers of the cloud. 4
2
 Layered Cloud Architectural
Development
• The application layer is formed with a collection of all
needed SW modules for SaaS applications.
—Service applications in this layer include daily office
management work, such as information retrieval, document
processing, and calendar and authentication services.
• The application layer is also heavily used by enterprises
in
business marketing and sales, consumer relationship
management (CRM), financial transactions, and supply
chain management.
—It should be noted that not all cloud services are restricted
to a single layer.
— Many applications may apply resources at mixed layers.
 Market-Oriented Cloud Architecture
• As consumers rely on cloud providers, they will
require a specific level of QoS to be maintained by
their providers:
— Cloud providers consider and meet the different QoS
parameters of each individual consumer as negotiated in
specific SLAs.
— To achieve this, the providers deploy market-
oriented resource management to regulate the supply and
demand of cloud resources to achieve market equilibrium
between supply and demand instead of traditional system-
centric resource management architecture.
—Figure 4.16 shows the high-level architecture for
market-oriented resource allocation in a
cloud computing environment.
Market-Oriented Cloud
Architecture
 Market-Oriented Cloud Architecture
• The market-oriented cloud is basically built with the
following entities:
—Users or brokers acting on user's behalf submit service
requests from anywhere in the world to the data
center.
—The SLA resource allocator acts as the interface between
the data center/cloud service provider and external
users/brokers.
—When a service request is first submitted, the service
request examiner interprets the request for QoS
requirements before to accept or reject the
request.
Market-Oriented Cloud Architecture
—The service request examiner ensures that there is
no overloading of resources whereby many service
requests cannot be fulfilled due to limited resources.
— It also needs the latest status information
regarding resource availability from the VM
Monitor mechanism) and workload processing
(/rom the Service Request Monitor mechanism) in
order to make resource allocation decisions
effectively.
—The VM Monitor mechanism keeps track of the
availability of VMs and their resource entitlements.
 Market-Oriented Cloud Architecture
• The dispatcher mechanism starts the execution of
accepted service requests on allocated VMs.
• The Service Request Monitor mechanism keeps
track of the execution progress of service
requests.
• Multiple VMs can be started and stopped on
demand on a single physical machine to meet
accepted service requests.
• Multiple VMs can concurrently run applications based
on different OS environments on a single physical
machine.
 Virtualization Support and Disaster
Recovery
• One of the distinguishing features of cloud
computing infrastructure is the use of system
virtualization:
— Virtualization of servers on a shared cluster can
consolidate web services.
• As the VMs are the containers of cloud services, the
provisioning tools will first find the corresponding physical
machines and deploy the VMs to those nodes before scheduling
the service to run on the virtual nodes:
— The user will not care about the computing resources
that are used for providing the services.
 Hardware Virtualization
• In many cloud computing systems, virtualization
SW is used to virtualization of the HW:
—Virtualization SW is a special kind of SW which
simulates the execution of HW and runs even
unmodified OSs.
—Virtualization software is also used as the platform for
developing new cloud applications that enable
developers to use any OSs and programming
environments they like.
—The development environment and deployment
environment can now be the same, which
eliminates some runtime problems.

50
 Hardware Virtualization
• As mentioned before, system virtualization SW is
considered the HW mechanism to run an
unmodified OS, usually on bare HW directly, on
top of SW.
• Table 4.4 lists some of the system virtualization
software in wide use at the time of this
writing.
—Currently, the VMs installed on a cloud computing
platform are mainly used for hosting third-
party programs.
—VMs provide flexible runtime services to free users from
worrying about the system environment.
51
Hardware Virtualization

50
Hardware Virtualization
 Hardware Virtualization
Users have full access to their own VMs, which
are completely separate from other users' VMs.
—Multiple VMs can be mounted on the same physical
server.
—Different VMs may run with different OSs.
• Also needs to establish the virtual disk storage
and virtual networks needed by the VMs:
—The virtualization is carried out by special servers
dedicated to generating the virtualized resource pool.
— The virtualized infrastructure is built with
virtualizing integrotiozt managers.
• These manager handle loads, resources, security, data, and
provisioning functions.
5
3
Cloud Architectural Design Challenges
• Challenge 1: Service Availability and Data Lock-in
Problem.
• Challenge 2: Data Privacy and Security Concerns.
• Challenge 3: Unpredictable Performance and
Bottlenecks.
• Challenge 4: Distributed Storage and Widespread
Software Bugs.
° Challenge 5: Cloud Scalability, Interoperability,
and Standardization.
• Challenge 6: Software Licensing and
Reputation
Sharing.
 Service Availability and Data Lock-in
Problem
• The management of a cloud service by a single company is
often the source of single points of failure.
• To achieve HA, one can consider using multiple cloud providers.
° Even if a company has multiple data centers located in different
geographic regions, it may have common SW infrastructure and
accounting systems.
• Therefore, using multiple cloud providers may provide
more protection from failures.
° Another availability obstacle is distributed denial of service
(DDoS) attacks. Some utility computing services offer SaaS
providers the opportunity to defend against DDoS attacks by using
quick scale-ups.
 Data Privacy and Security
• Cloud offerings through public networks, exposing the
Concerns
system to more attacks.
• Many attacks can be overcome with technologies such as
encrypted storage, virtual LANs, and network
middleboxes (e.g., firewalls, packet filters).
— For example, you could encrypt your data before placing it in a
cloud.
• Impose laws requiring SaaS providers to keep customer data
and copyrighted material within national boundaries.
—In a cloud environment, attacks may result from hypervisor
malware, guesr hopping and fiJocf‹ing, or VM rootkits.
• In general, passive attacks steal sensitive data or
passwords.
• Active attacks may manipulate kernel data structures which
will cause major damage to cloud servers.
 Unpredictable Performance and
Bottlenecks
• Multiple VMs can share CPUs and main memory
in cloud computing, but I/O sharing is
problematic:
—This issue is due to the problem of I/O interference
between VMs.
• One solution is to improve I/O architectures
and
operating systems to efficiently virtualizes
interrupts and I/O channels.
—Internet applications continue to become more data-
intensive.
—If we assume applications to be “pulled apart” across the
boundaries of clouds, this may complicate data
placement and transport.
 Distributed Storage and Widespread
Software Bugs
• The design of efficient distributed storage area
networks (SANs) is based on the database and
storage system of the cloud network.
• Data consistence checking in SAN-connected data
centers is a major challenge in cloud
computing.
• Large-scale distributed bugs cause debugging and that
must occur at the data centers.
—No data center will provide such a convenience.
• One solution may be a reliance on using VMs in
cloud computing.
—The level of virtualization may make it possible to capture
valuable information.
 Cloud Scalability, Interoperability, and
Standardization
• Computation in a cloud is different depending on
virtualization level.
—Google App Engine(GAE) automatically scales in response
to load increases and decreases; users are charged
by the cycles used.
—Amazon Web Service (AWS) charges by the hour for the
number of VM instances used, even if the machine
is idle.
• The opportunity here is to scale quickly up and
down in response to load variation, in order to
save money, but without violating SLAs.
 Cloud Scalability, Interoperability, and
Standardization
• Open Virtualization Format (OVF) describes an open,
secure, porto6/e, efficient, and extensible format for
the packaging and distribution of VMs.
—It also defines a format for distributing SW to be
deployed in VMs.
—This VM format does not rely on the use of a specific
host platform, virtualization platform, or guest OS.
• The approach is to address virtual platform-agnostic
packaging with certification and integrity of packaged
software.
—The package supports virtual appliances to span
more than one VM.
 Software Licensing and Reputation
Sharing
• Many cloud computing providers originally relied on
open source SW because the licensing model for
commercial software is not ideal for utility
computing.
• The primary opportunity is either for open
source to remain popular or simply for commercial
software companies to change their licensing
structure to better fit cloud computing.
• One can consider using both pay-for-use and
bulk- use licensing schemes to widen the
business coverage.
 Public Clouds and Service
Offerings
• Cloud services are demanded by computing and IT
administrators, software vendors, and end users (Figure 4.19
introduces five levels of cloud players):
—At the top level, individual users and organizational
users
demand very different services.
—The application providers at the SaaS level serve
mainly individual users.
—Most business organizations are serviced by IaaS and PaaS
providers.
— The IaaS provide compute, storage, and
communication resources to both applications and
organizational users.
— The cloud environment is defined by the PaaS or
platform providers.
° Note that the platform providers support both
infrastructure services and organizational users directly.
Public Clouds and Service Offerings
 Public Clouds and Service
• Offerings
Cloud services rely on new advances in machine
virtualization, SOA, grid infrastructure management, and
power efficiency.
— Consumers purchase such services in the form of IaaS, PaaS,
or SaaS.
• The cloud industry leverages the growing demand by
many enterprises and business users to outsource their
computing and storage jobs to professional providers.
— The provider service charges are often much lower
than the cost for users to replace their obsolete servers
frequently.
• Table 4.5 summarizes the profiles of five major cloud
providers by 2010 standards.
 Google App Engine (GAE)
• Google has the world’s largest search engine facilities.
The company has extensive experience in massive data
processing that has led to new insights into data-center
design and novel programming models that scale to
incredible sizes.
• Google has hundreds of data centers and has installed
more than 460,000 servers worldwide. For example, 200
Google data centers are used at one time for a number
of cloud applications. Data items are stored in text,
images, and video and are replicated to tolerate faults or
failures. Here we discuss Google’s App Engine (GAE)
which offers a PaaS platform supporting various cloud
and web applications.
 Google App Engine (GAE)
 Google Cloud Infrastructure
Leveraging the large number of data centers it operates. For
example, Google pioneered cloud services in Gmail, Google Docs,
and Google Earth, among other applications. These applications
can support a large number of users simultaneously with HA.
Notable technology achievements include the Google File System
(GFS), MapReduce, BigTable, and Chubby. In 2008, Google
announced the GAE web application platform which is becoming a
common platform for many small cloud service providers. This
platform specializes in supporting scalable (elastic) web
applications. GAE enables users to run their applications on a
large number of data centers associated with Google’s search
engine operations.
Google App Engine (GAE)
 Google App Engine (GAE)
 GAE Architecture
Major building blocks of the Google cloud platform which has
been used to deliver the cloud services highlighted earlier. GFS
is used for storing large amounts of data. MapReduce is for use
in application program development. Chubby is used for
distributed application lock services. BigTable offers a storage
service for accessing structured data.
Third-party application providers can use GAE to build cloud applications
for providing services.
Google is one of the larger cloud application providers, Chubby for
distributed application lock services, and BigTable as a storage service for
accessing structural or semistructural data. With these building blocks,
Google has built many cloud applications. Chubby for distributed locks can
also run in the clusters. GAE runs the user program on Google’s
infrastructure.
 Google App Engine (GAE)
 Functional Modules of
TheGAE
GAE platform comprises the following five major components. The GAE
is not an infrastructure platform, but rather an application development
platform for users. We describe the component functionalities separately.

a. The datastore offers object-oriented, distributed, structured data storage services based on BigTable
techniques. The datastore secures data management operations.

b. The application runtime environment offers a platform for scalable web programming and execution. It
supports two development languages: Python and Java.

c. The software development kit (SDK) is used for local application development. The SDK allows users to
execute test runs of local applications and upload application code.

d. The administration console is used for easy management of user application development cycles, instead
of for physical resource management.

e. The GAE web service infrastructure provides special interfaces to guarantee flexible use and management
of storage and network resources by GAE.
 Google App Engine (GAE)
 GAE Applications

GAE applications include the Google Search Engine, Google Docs, Google Earth, and Gmail. These applications
can support large numbers of users simultaneously. Users can interact with Google applications via the web
interface provided by each application. Third-party application providers can use GAE to build cloud applications
for providing services. The applications are all run in the Google data centers. Inside each data center, there might
be thousands of server nodes to form different clusters. (See the previous section.) Each cluster can run
multipurpose servers.

GAE supports many web applications. One is a storage service to store application-specific data in the Google
infrastructure. The data can be persistently stored in the backend storage server while still providing the facility for
queries, sorting, and even transactions similar to traditional database systems. GAE also provides Google-
specific services, such as the Gmail account service (which is the login service, that is, applications can use the
Gmail account directly). This can eliminate the tedious work of building customized user management
components in web applications. Thus, web applications built on top of GAE can use the APIs authenticating
users and sending e-mail using Google accounts.
 Amazon Web Services
• VMs for Resource Sharing: VMs can share computing
resources flexibly and safely.

• Amazon's Leadership: Amazon is a leader in public cloud

services, applying the IaaS model.

• AWS Architecture: EC2 provides virtualized platforms for

hosting VMs, S3 offers object-oriented storage, and EBS
provides block storage.

• SQS: Simple Queue Service ensures reliable messaging

between processes.

• SOAP Access: Users can access objects through SOAP

with browsers or client programs.
 Amazon Web Services
• Service Offerings: AWS offers various services
summarized in Table 4.6, with details in Chapter 6.

• Queuing and Notification: Amazon provides SQS and

SNS for queuing and notification services.

• Flexible Platform: Amazon offers a flexible cloud

computing platform for developers.

• Business Support: Small and medium-sized companies

can use AWS to service large numbers of Internet users.

• ELB and CloudWatch: ELB distributes traffic across EC2

instances, and CloudWatch monitors resources.
 Amazon Web Services
• RDS and Elastic MapReduce: Amazon offers RDS and
Elastic MapReduce, equivalent to Hadoop on EC2.

• AWS Import/Export: Allows shipping large volumes of

data to and from EC2 using physical disks.

• CloudFront: Implements a content distribution network.

• Amazon DevPay: Provides online billing and account

management for businesses.

• FPS: Allows developers to charge customers using

Amazon's payment system.
 Amazon Web Services
• FWS: Merchants can use Amazon's fulfillment capabilities
through a web service interface.

• MPI Clusters: Amazon offers MPI clusters and cluster

compute instances with hardware-assisted virtualization.
Amazon Web Services
Amazon Web Services
 Microsoft Windows Azure
• In 2008, Microsoft launched a Windows Azure platform to
meet the challenges in cloud computing.
• This platform is built over Microsoft data centers.
• Figure 4.22 shows the overall architecture of Microsoft's cloud
platform.
• The platform is divided into three major component
platforms:
— Windows Azure offers a cloud platform built on Windows
OS and based on Microsoft virtualization technology.
— Applications are installed on VMs deployed on the
data-center servers.
— Azure manages all servers, storage, and network resources
of the data center.
— On top of the infrastructure are the various services for
building different cloud applications.
 Microsoft Windows Azure
• Cloud-level services provided by the Azure platform are
introduced below:
—Live service: Users can visit Microsoft Live applications and
access multiple machines concurrently.
— .NET service: This package supports application development
on local hosts and execution on cloud machines.
— SQL Azure: This function makes it easier for users to
visit and use the relational database with the SQL server in
the cloud.
— SharePoint service: This provides a scalable and
manageable platform for users to develop business
applications in upgraded web services.
— Dynamic CRM service: This provides SW developers a business
platform in managing CRM applications in financing,
marketing, sales, etc.
Microsoft Windows Azure
 Extended Cloud Computing Services
• Figure 4.23 shows six layers of cloud services, ranging from
hardware, network, and collocation to infrastructure, platform,
and software applications.
• We already introduced the top three service layers as
SaaS, PaaS, and IaaS, respectively.
• The cloud platform provides PaaS, which sits on top of the
IaaS infrastructure.
• The top layer offers SaaS.
• These must be implemented on the cloud platforms provided.
• Although the three basic models are dissimilar in usage, as
shown in Table 4.7, they are built one on top of another.
Extended Cloud Computing Services
Extended Cloud Computing Services
 Extended Cloud Computing Services
• Cloud Service Tasks and Trends

• Cloud services are introduced in five layers.

• SaaS applications are mostly for business applications like CRM.
• PaaS is provided by companies like Google, Salesforce.com, and
Facebook.
• IaaS is provided by Amazon, Windows Azure, and RackRack.
• Network cloud services are provided by AT&T, Qwest, and AboveNet.
• Vertical cloud services refer to mutually supportive cloud services and
often involve cloud mashup.
 Extended Cloud Computing Services
• Software Stack for Cloud Computing
• Cloud computing clusters have software stacks built from scratch to meet
rigorous goals.

• Developers design systems to meet high throughput, high availability, and

fault tolerance.

• Cloud computing software stacks are structured in layers, each with its
own purpose.

• Platforms for cloud services can be physical or virtual servers, providing

flexibility.

• The software layer on top of the platform stores massive amounts of data,
similar to a file system.

• Other layers include database storage, programming for large-scale

clusters, and data query language support.
 Extended Cloud Computing Services
• Runtime Support Services

• Runtime supporting services in cloud computing include cluster monitoring

and job management systems.
• The scheduler queues tasks and assigns them to processing nodes based on
availability.
• Distributed schedulers support cloud applications, such as those using
MapReduce.
• Runtime support is needed for browser-initiated applications used by many
cloud customers.
• The SaaS model provides software as a service, eliminating upfront
investments for customers.
• Customer data is stored in either vendor proprietary or publicly hosted clouds
supporting PaaS and IaaS.
 Extended Cloud Computing Services
• Cloud Service Tasks and Trends

• Cloud services are introduced in five layers.

• SaaS applications are mostly for business applications like CRM.
• PaaS is provided by companies like Google, Salesforce.com, and
Facebook.
• IaaS is provided by Amazon, Windows Azure, and RackRack.
• Network cloud services are provided by AT&T, Qwest, and AboveNet.
• Vertical cloud services refer to mutually supportive cloud services and
often involve cloud mashup.
 Virtual Machine Creation and Management
• Resource management of independent service jobs is a key issue in cloud infrastructure
management.
• Cloud-loading experiments on the French Grid'5000 system illustrate VM creation and
management.
• VM managers provide a public API for users to submit and control VMs.
 Virtual Machine Creation and Management
 Independent Service Management
• Independent services request facilities to execute many unrelated tasks. Commonly, the
APIs provided are some web services that the developer can use conveniently. In Amazon
cloud computing infrastructure, SQS is constructed for providing a reliable communication
service between different providers. Even the endpoint does not run while another entity
has posted a message in SQS. By using independent service providers, the cloud
applications can run different services at the same time. Some other services are used for
providing data other than the compute or storage services.

 Running Third-Party Applications

• Cloud platforms support building applications by third-party providers or programmers
using Web 2.0 forms and APIs in the form of services.

• Cloud computing platforms provide extra capabilities for accessing backend services or
underlying data, with examples like GAE and Microsoft Azure.

• Various application engines, such as WebSphere for BlueCloud and any engine that can
run in EC2 VM instances, are used for developing web applications.
 Virtual Machine Creation and Management
 Virtual Machine Manager
• The VM manager links the gateway and resources, using virtualization technology to
manage VMs on physical resources.
• The Melbourne group has developed VM managers for OpenNebula, Amazon EC2, and
French Grid'5000.
• Users submit VMs on physical machines using hypervisors like Xen, and the VM manager
handles deployment on grids and IaaS providers.

 Virtual Machine Templates

A VM template is analogous to a computer’s configuration and contains a description for a
VM with the following static information:
• The number of cores or processors to be assigned to the VM
• The amount of memory the VM requires
• The kernel used to boot the VM’s operating system
• The disk image containing the VM’s file system
• The price per hour of using a VM
 Virtual Machine Creation and Management
 Virtual Machine Templates
the VMM generates a descriptor from the template. This
descriptor contains the same fields as the template and additional information related to a
specific VM instance. Typically the additional information includes: The number of cores or
processors to be assigned to the VM

• The disk image that contains the VM’s file system

• The address of the physical machine hosting the VM
• The VM’s network configuration
• The required information for deployment on an IaaS provider

Network information is not needed, because Amazon EC2 automatically assigns a public IP to
the instances. The IGG works with a repository of VM templates, called the VM template
directory.
 Virtual Machine Creation and Management
 Distributed VM Management

• The local gateway tries to obtain resources from the underlying VIEs. When this is
impossible, the local gateway starts a negotiation with any remote gateways to fulfill the
request. When a gateway schedules the VMs, it sends the VM access information to the
requester gateway.
 Virtual Machine Creation and Management
 Global Exchange and Resource
• Cloud infrastructure providers have established data centers in multiple geographical
locations to provide redundancy and ensure reliability.
• Amazon expects cloud customers to express a preference for hosting locations, but does not
provide automatic scaling across multiple data centers.
• Cloud customers may struggle to determine the best hosting location and meet QoS
expectations for consumers from multiple locations.
• The Cloud bus Project at the University of Melbourne proposed InterCloud architecture for
seamless federation of data centers and dynamic scaling of applications.
 Virtual Machine Creation and Management
 Global Exchange and Resource

• InterCloud architecture allows cloud

providers to expand or resize provisioning
capabilities based on workload demands
and operate as part of a market-driven
resource leasing federation.
• Client brokering and coordinator services
support utility-driven federation of clouds,
application scheduling, resource
allocation, and workload migration.
• The Cloud Exchange (CEx) acts as a
market maker, supporting trading of cloud
services based on competitive economic
models and ensuring secure financial
transactions.