Inter-cloud Resource Management

Cloud resource management and intercloud resource exchange schemes are

reviewed as follows.
1. Extended Cloud Computing Services
2. Cloud Service Tasks and Trends
3. Software Stack for Cloud Computing
4. Runtime Support Services

UNIT-III CLOUD INFRASTRUCTURE 19

1.Extended Cloud Computing Services

UNIT-III CLOUD INFRASTRUCTURE 20

1.Extended Cloud Computing Services ….
 Figure shows six layers of cloud services, ranging from hardware, network, and
collocation to infrastructure, platform, and software applications.
 The cloud platform provides PaaS, which sits on top of the IaaS infrastructure. The
top layer offers SaaS.
 The bottom three layers are more related to physical requirements. The bottom most
layer provides Hardware as a Service (HaaS).
 The next layer is for interconnecting all the hardware components, and is simply
called Network as a Service (NaaS). Virtual LANs fall within the scope of NaaS.
 The next layer above NaaS offers Location as a Service (LaaS), which provides a
collocation service to house, power, and secure all the physical hardware and
network resources.
 The cloud infrastructure layer can be further subdivided as Data as a Service (DaaS)
and Communication as a Service (CaaS) in addition to compute and storage in IaaS.

UNIT-III CLOUD INFRASTRUCTURE 21

1.Extended Cloud Computing Services ….
The cloud players are divided into three classes:
a) cloud service providers and IT administrators,
b) software developers or vendors, and
c) end users or business users.
 These cloud players vary in their roles under the IaaS, PaaS, and SaaS models.
 From the software vendors’ perspective, application performance on a given cloud
platform is most important.
 From the providers’ perspective, cloud infrastructure performance is the primary
concern.
 From the end users’ perspective, the quality of services, including security, is the most
important.

UNIT-III CLOUD INFRASTRUCTURE 22

 2.Cloud Service Tasks and Trends
 The top layer is for SaaS applications, mostly used for business applications.
 The approach is to widen market coverage by investigating customer behaviours and
revealing opportunities by statistical analysis.
 SaaS tools also apply to distributed collaboration, and financial and human resources
management.
 PaaS is provided by Google, Salesforce.com, and Facebook
 IaaS is provided by Amazon, Windows Azure, and Rackspace
 Collocation services require multiple cloud providers to work together to support supply
chains in manufacturing.
 Network cloud services provide communications such as those by AT&T, Qwest, and
Above Net.

UNIT-III CLOUD INFRASTRUCTURE 23

 3. Software Stack for Cloud Computing
 The overall software stack structure of cloud computing software can be viewed
as layers.
 Each layer has its own purpose and provides the interface for the upper layers just
as the traditional software stack does.
 However, the lower layers are not completely transparent to the upper layers.
 The platform for running cloud computing services can be either physical servers
or virtual servers.
 By using VMs, the platform can be flexible, that is, the running services are not
bound to specific hardware platforms. This brings flexibility to cloud computing
platforms.

UNIT-III CLOUD INFRASTRUCTURE 24

3. Software Stack for Cloud Computing….

 The software layer on top of the platform is the layer for storing massive
amounts of data.
 This layer acts like the file system in a traditional single machine.
 Other layers running on top of the file system are the layers for executing cloud
computing applications.
 They include the database storage system, programming for large-scale
clusters, and data query language support.

UNIT-III CLOUD INFRASTRUCTURE 25

 4. Runtime Support Services
 As in a cluster environment, there are also some runtime supporting services in the cloud
computing environment.
 Cluster monitoring is used to collect the runtime status of the entire cluster. One of the most
important facilities is the cluster job management system.
 The scheduler queues the tasks submitted to the whole cluster and assigns the tasks to the
processing nodes according to node availability.
 The distributed scheduler for the cloud application has special characteristics that can
support cloud applications, such as scheduling the programs written in MapReduce style.
 The runtime support system keeps the cloud cluster working properly with high efficiency.
 Runtime support is software needed in browser-initiated applications applied by thousands
of cloud customers.
 The SaaS model provides the software applications as a service, rather than letting users
purchase the software

UNIT-III CLOUD INFRASTRUCTURE 26

 Resource Provisioning and Platform Deployment

In this section, will discuss techniques to provision computer resources or VMs

and storage allocation schemes.
1. Provisioning of Compute Resources
2.Resource Provisioning Methods
3. Dynamic Resource Deployment
4. Provisioning of Storage Resources

UNIT-III CLOUD INFRASTRUCTURE 27

 1. Provisioning of Compute Resources
 Providers offer cloud services by signing SLAs with end users.
 The SLAs must commit sufficient resources such as CPU, memory, and bandwidth that the
user can use for a preset period.
 Under -provisioning of resources will lead to broken SLAs and penalties.
 Overprovisioning of resources will lead to resource underutilization, and consequently, a
decrease in revenue for the provider.
 Deploying an autonomous system to efficiently provision resources to users is a
challenging problem.
 The difficulty comes from the unpredictability of consumer demand, software and
hardware failures, heterogeneity of services, power management, and conflicts in signed
SLAs between consumers and service providers.

UNIT-III CLOUD INFRASTRUCTURE 28

1. Provisioning of Compute Resources….
 Efficient VM provisioning depends on the cloud architecture and management of cloud
infrastructures.
 Resource provisioning schemes also demand fast discovery of services and data in
cloud computing infrastructures.
 In a virtualized cluster of servers, this demands efficient installation of VMs, live VM
migration, and fast recovery from failures.
 To deploy VMs, users treat them as physical hosts with customized operating systems for
specific applications.
 The provider should offer resource economic services. Power-efficient schemes for
caching, query processing, and thermal management are mandatory due to increasing
energy waste by heat dissipation from data centers.

UNIT-III CLOUD INFRASTRUCTURE 29

 2. Resource Provisioning Methods
There are three cases of static cloud resource provisioning policies.
In case (a) : overprovisioning with the peak load causes heavy resource waste (shaded area).
In case(b) : under provisioning (along the capacity line) of resources results in losses by both
user and provider
In case (c ) : the constant provisioning of resources with fixed capacity to a declining user
demand could result in even worse resource waste.

Three resource-provisioning methods are presented here:

a) Demand-driven method
b) Event-driven method
c) Popularity-driven method

UNIT-III CLOUD INFRASTRUCTURE 30

2. Resource Provisioning Methods….

(a) Demand-driven method :

 This method adds or removes computing instances based on the current utilization
level of the allocated resources.
 In general, when a resource has surpassed a threshold for a certain amount of time,
the scheme increases that resource based on demand.
 When a resource is below a threshold for a certain amount of time, that resource
could be decreased accordingly.
 Amazon implements such an auto-scale feature in its EC2 platform.

UNIT-III CLOUD INFRASTRUCTURE 31

2. Resource Provisioning Methods….
(b) Event-driven method :
 This scheme adds or removes machine instances based on a specific time event.
 The scheme works better for seasonal or predicted events such as Christmas time
in the West and the Lunar New Year in the East. (Festival and holidays)
 During these events, the number of users grows before the event period and then
decreases during the event period.
 This scheme anticipates peak traffic before it happens. The method results in a
minimal loss of QoS, if the event is predicted correctly.
(C ) Popularity-driven method :
 In this method, the Internet searches for popularity of certain applications and
creates the instances by popularity demand.
 The scheme anticipates increased traffic with popularity. Again, the scheme has a
minimal loss of QoS, if the predicted popularity is correct Resources may be
wasted if traffic does not occur as expected.
UNIT-III CLOUD INFRASTRUCTURE 32
 3.Dynamic Resource Deployment
 The cloud uses VMs as building blocks to create an execution environment across
multiple resource sites.
 Dynamic resource deployment can be implemented to achieve scalability in
performance.
 The InterGrid is a Java-implemented software system that lets users create
execution cloud environments on top of all participating grid resources.
 Peering arrangements established between gateways enable the allocation of
resources from multiple grids to establish the execution environment.

UNIT-III CLOUD INFRASTRUCTURE 33

3.Dynamic Resource Deployment…
In figure, a scenario is illustrated by which an InterGrid gateway (IGG) allocates
resources from a local cluster to deploy applications in three steps:
(1) requesting the VMs,
(2) enacting the leases, and
(3) deploying the VMs as requested.

Under peak demand, this IGG interacts with another IGG that can allocate resources
from a cloud computing provider.

UNIT-III CLOUD INFRASTRUCTURE 34

UNIT-III CLOUD INFRASTRUCTURE 35
3. Dynamic Resource Deployment…

 Grid has predefined peering arrangements with other grids, which the IGG
manages. Through multiple IGGs, the system coordinates the use of InterGrid
resources.
 An IGG is aware of the peering terms with other grids, selects suitable grids that
can provide the required resources, and replies to requests from other IGGs.
 An IGG can also allocate resources from a cloud provider.
 The cloud system creates a virtual environment to help users deploy their
applications. These applications use the distributed grid resources.

UNIT-III CLOUD INFRASTRUCTURE 36

3.Dynamic Resource Deployment…

 The InterGrid allocates and provides a distributed virtual environment (DVE).

 This is a virtual cluster of VMs that runs isolated from other virtual clusters.
 A component called the DVE manager performs resource allocation and
management on behalf of specific user applications.
 The core component of the IGG is a scheduler for implementing provisioning
policies and peering with other gateways.
 The communication component provides an asynchronous message-passing
mechanism. Received messages are handled in parallel by a thread pool.

UNIT-III CLOUD INFRASTRUCTURE 37

 4.Provisioning of Storage Resources
 The data storage layer is built on top of the physical or virtual servers provider. The
service can be accessed anywhere in the world.
 One example is e-mail systems. A typical large e-mail system might have millions of
users and each user can have thousands of e-mails and consume multiple gigabytes of
disk space.
 Another example is a web searching application. In storage technologies, hard disk
drives may be augmented with solid-state drives in the future.
 The biggest barriers to adopting flash memory in data centers have been price,
capacity and to some extent, a lack of sophisticated query-processing techniques.

UNIT-III CLOUD INFRASTRUCTURE 38

4.Provisioning of Storage Resources…
 Distributed file system is very important for storing large-scale data. However, other
forms of data storage also exist.
 Some data does not need the namespace of a tree structure file system, and instead,
databases are built with stored data files. In cloud computing, another form of data
storage is (Key, Value) pairs.
 Many cloud computing companies have developed large-scale data storage systems to
keep huge amount of data collected every day.
For example :
• Google’s GFS stores web data and other data, such as geographic data for Google Earth.
• A similar system from the open source community is the Hadoop Distributed File System (HDFS)
for Apache. Hadoop is the open source implementation of Google’s cloud computing
infrastructure.
• Similar systems include Microsoft’s Cosmos file system for the cloud.

UNIT-III CLOUD INFRASTRUCTURE 39

4.Provisioning of Storage Resources …
 The main purpose is to store the data in structural or semi-structural ways so that
application developers can use it easily and build their applications rapidly.
 Web pages are an example of semi structural data in HTML format. For example :
applications might want to process the information contained in a web page.
 Traditional databases will meet the performance bottleneck while the system is
expanded to a larger scale.
 However, some real applications do not need such strong consistency. The scale of
such databases can be quite large.
 Typical cloud databases include BigTable from Google, SimpleDB from Amazon, and
the SQL service from Microsoft Azure.

UNIT-III CLOUD INFRASTRUCTURE 40

 Global Exchange of Cloud Resources
 To support a large number of application service consumers from around the world,
cloud infrastructure providers (i.e., IaaS providers) have established data centres in
multiple geographical locations to provide redundancy and ensure reliability in case of
site failures.
 Amazon does not provide seamless/automatic mechanisms for scaling its hosted
services across multiple geographically distributed data centres.
 Due to this, first it is difficult for cloud customers to determine in advance the best
location for hosting their services as they may not know the origin of consumers and
their services.
 Second, SaaS providers may not be able to meet the QoS expectations of their service
consumers originating from multiple geographical locations.

UNIT-III CLOUD INFRASTRUCTURE 41

Global Exchange of Cloud Resources…
 In addition, no single cloud infrastructure provider will be able to establish its data
centres at all possible locations throughout the world.
 This necessitates building mechanisms for seamless federation of data centers of a
cloud provider or providers supporting dynamic scaling of applications across multiple
domains in order to meet QoS targets of cloud customers.
 Hence, they would like to make use of services of multiple cloud infrastructure service
providers who can provide better support for their specific consumer needs.
 To realize this, InterCloud architecture is developed for supporting brokering and
exchange of cloud resources for scaling applications across multiple clouds.
 Figure illustrates the components of InterCloud architecture.

UNIT-III CLOUD INFRASTRUCTURE 42

InterCloud Architecture

UNIT-III CLOUD INFRASTRUCTURE 43

InterCloud Architecture…
 They consist of client brokering and coordinator services that support utility-driven
federation of clouds: application scheduling, resource allocation, and migration of
workloads.
 The architecture couples the administratively and topologically distributed storage
and compute capabilities of clouds as part of a single resource leasing abstraction.
 The system will ease the cross-domain capability integration for on-demand,
flexible, energy-efficient, and reliable access to the infrastructure based on
virtualization technology.
 The Cloud Exchange (CEx) acts as a market maker for bringing together service
producers and consumers with fitting offers.

UNIT-III CLOUD INFRASTRUCTURE 44

InterCloud Architecture…
 It aggregates the infrastructure demands from application brokers and evaluates
them against the available supply currently published by the cloud coordinators.
 It supports trading of cloud services based on competitive economic models such
as commodity markets and auctions.
 Such markets enable services to be commoditized, and thus will pave the way for
creation of dynamic market infrastructure for trading based on SLAs.
 An SLA specifies the details of the service to be provided in terms of metrics
agreed upon by all parties, and incentives and penalties for meeting and violating
the expectations, respectively.
 The availability of a banking system within the market ensures that financial
transactions pertaining to SLAs between participants are carried out in a secure
and dependable environment.

UNIT-III CLOUD INFRASTRUCTURE 45