Cloud Analytics and VDI

Unit 2
Cloud Analytics, Testing Under Control,
Virtual Desktop Infrastructure
Dr. Neelesh Kumar Jain
Professor

Cloud Analytics
Cloud analytics is a service model where one or more components of
analytics are implemented in the cloud. These services could be part
of a hybrid model, where some components are on-premise, or fully
in the cloud. The cloud model allows organizations to scale analytics
capabilities as their company grows. It also removes the burden of
on-premise management and implementation. This service model is
a growing aspect of modern business intelligence systems today.

Gartner defined six elements of analytics as:
•Data sources
These are the original sources of data which could include ERPs,
CRMs, social media data, or website usage data. An example of a
cloud-based data source would be Twitter sentiment data.
•Data models
Cloud-based data models make sense of and standardize how
data points are related to each other. These are typically created
with structured data types.
•Processing applications
These applications process large volumes of big data, as it’s
ingested into a data warehouse. Hadoop is a popular application
for data processing.
Components of Cloud Analytics

•Computing power
Companies need raw computing power at scale to
ingest, structure, clean, analyze, and serve business
data.
•Analytic models
These mathematical models are closed functions
used to predict outcomes and require strong
computing power to create.
•Sharing and/or storage of
Data warehouses as a service enable organizations
to quickly implement a modern analytics
architecture and easily scale.
Components of Cloud Analytics

Benefits of Cloud Analytics
• Enterprise data consolidation
Large enterprises have many disparate data sources, and it’s difficult
to see how all the moving parts of an organization are working
together if they’re in different places. A cloud implementation can
provide a data warehouse that’s accessible to those who need the
data. Companies can easily ensure data governance so only those
who need the data get it. Another advantage of consolidation is the
ability to use online services to perform data mining and advanced
analytics to create prediction models updated in real time.

• Ease of access
• Data in the cloud can be accessed by both employees and external
stakeholders, and governance controls can be put in place to control
access to the right people. Managing access from disparate data
sources requires more resources to manage internally and slows
down innovation and insights.

• Sharing and collaboration
• Increased ease of access and data consolidation lead to more
sharing and collaboration between employees, which is why cloud
analytics is a good fit for global companies. Employees can easily
transfer files and collaborate in real time when they view analytics in
the cloud from anywhere in the world. This is also conducive to the
growing trend of a telecommuting work culture. Data
discovery becomes an everyday part of the culture when cloud
analytics is implemented within a BI system.

• Reduced operating costs
• Setting up an in-house analytics solution can be extremely costly,
especially for smaller organizations who may not have the internal
skillsets to do so. With cloud analytics, organizations don’t need to
purchase hardware and provide continuous support, which can be
very demanding and creates vulnerability if not properly executed.
There are also ongoing upgrades which need to occur and can
create unnecessary downtime. A cloud solution will take this burden
off an organization’s hands so they can focus on their core
competency.

• Scalability
• It’s also easier to scale up capacity as the business grows, as the
organization can simply increase its number of subscriptions as
opposed to purchasing new hardware. It also ensures systems scale
up accordingly if there is a sudden increase in demand for the
analytics systems.

Challenges of Cloud Analytics
• Performance: For enterprises where 100% uptime is needed, a hybrid approach to
cloud analytics may be the best approach.
• Data security: According to RightScale, data security is the number one challenge cited
by corporate cloud users. They fear data loss and leakage, and cloud implementation
can create some vulnerabilities.
• Finding the right skillsets for the job: Companies have a hard time ensuring they have
the right skillsets to build and manage a cloud analytics operation. The challenge is
training and hiring to keep up with changing technology.
• Managing cost: The on-demand nature of cloud analytics allows organizations to scale
as needed, but companies often underestimate how much they will use their cloud
analytics capabilities.
• Migration: Migrating legacy systems such as data warehouses to the cloud can be
time-consuming and expensive. There’s also a risk of data loss if the migration is not
done properly.

Cloud Testing
• Cloud testing is a subset of software testing in which simulated, real-
world Web traffic is used to test cloud-based Web applications. Cloud
testing also verifies and validates specific cloud functions, including
redundancy and performance scalability.

Type of Testing in Cloud
• The whole cloud testing is segmented into four main categories
• Testing of the whole cloud: The cloud is viewed as a whole entity and
based on its features testing is carried out. Cloud and SaaS vendors, as well
as end users, are interested in carrying out this type of testing
• Testing within a cloud: By checking each of its internal features, testing is
carried out. Only cloud vendors can perform this type of testing
• Testing across cloud: Testing is carried out on different types of cloud-like
private, public and hybrid clouds
• SaaS testing in cloud: Functional and non-functional testing is carried out
on the basis of application requirements

Cloud testing focuses on the core
components like:
• Application: It covers testing of functions, end-to-end business
workflows, data security, browser compatibility, etc.
• Network: It includes testing various network bandwidths, protocols
and successful transfer of data through networks.
• Infrastructure: It covers disaster recovery test, backups, secure
connection, and storage policies. The infrastructure needs to be
validated for regulatory compliances

Other Testing types in Cloud includes
• Performance
• Availability
• Compliance
• Security
• Scalability
• Multi-tenancy
• Live upgrade testing

• VDI (Virtual Desktop Infrastructure) is a technology used to create a
virtualized desktop environment on a remote server setup. VDI segments
the servers into various virtual desktops which the users can access
remotely through their devices. These virtual desktops are hosted on
Virtual Machines (VM) that are controlled through management software.
• For instance, as an owner of an accounting firm, imagine the inconvenience
you face while installing and configuring all those desktop PCs and the
maintenance and upgrade requirement that keep coming with time.
• With VDI, as the desktops are virtual, you can access them even with a thin
client.

• As far as the users are concerned, VDI gives you the freedom of
accessing your desktop from anywhere at any time through a VDI
client software.
• VDI can be classified as persistent and non-persistent. Persistent VDI
is customized for a personal user, and the user can log in to the same
desktop each time. Non-persistent VDI consists of desktops that
revert to their initial state after the user logs out.

Benefits of VDI
1. Access
2. Security
3. Device Portability
4. Data Center Facilities
5. Cost Reduction

VDI Basic Components
1. Virtualization
• Virtualization is the technology that divides the system architecture
into different layers. Before virtualization, the hardware was bound to
the operating system (OS) at the time of installation. Thus, in the case
of a hardware failure, the OS also crashed, and you would lose all the
data.
• Through virtualization, the OS and the underlying hardware are
separated by software called hypervisor. You can install multiple
operating systems on a hypervisor installed server.

2. Hypervisor
• Hypervisor is software that separates the operating system from the
underlying hardware. The hypervisor creates a virtualized environment in
which the hardware can be divided into multiple virtual machines (VMs).
Each virtual machine can have its unique configuration, OS, and
applications.
• In VDI, the hypervisor creates desktop instances in these VMs. Each
desktop instance can act as a separate desktop and can be provisioned to
the users.
The High Availability (HA) function in the hypervisor also lets it connect to
multiple servers. Hence, even if a physical server fails, your desktop
instance is moved to another server almost instantaneously.

3. Connection Broker
• Connection broker is a software program that connects the users to
the desktop instances. It is also responsible for the authentication of
the users and sending them to their desktop instances.
• The connection broker also keeps track of active and inactive
desktops. When a user sends a request to connect to a desktop, it
provides the user with an idle desktop instance. When a user
disconnects desktop, it updates the status to inactive.

4. Desktop Pools
• Desktops pools are a group of similar desktops that can be configured
according to a specific function. For instance, departments like
accounting and IT in an office might use desktops with different
applications and configuration. The accounting department may need
applications like Sage or QuickBooks, which may not be required for
IT.
You can create a desktop pool for these departments with the similar
configuration

5. Application Virtualization
• Application Virtualization is the technology used to create a
virtualized application image and replicate it to all the virtual
desktops in a desktop pool. It makes application deployment easy and
hassle-free.
• VMware ThinApp is an example of that. It creates an executable file of
the application by capturing pre and post images of the system before
and after installation of the software. You can use this file in any
system without going through the installation process repeatedly.

Cloud Analytics and VDI

Recommandé

Recommandé

Contenu connexe

Tendances

Tendances (20)

Similaire à Cloud Analytics and VDI

Similaire à Cloud Analytics and VDI (20)

Plus de Dr Neelesh Jain

Plus de Dr Neelesh Jain (13)

Dernier

Dernier (20)

Cloud Analytics and VDI