ETCA_5

Manage to Success in Cloud Computing
Projects

Praveena Parameswaran - Technology Architect, Infosys Technologies Ltd, India

Contents
1. Abstract ......................................................................................................................................... 4
2. Introduction ................................................................................................................................... 4
3. What is Cloud Computing ........................................................................................................... 5
4. Conclusion .................................................................................................................................. 10
5. References .................................................................................................................................. 11
6. Author’s Profile: .......................................................................................................................... 11

3 Page

1. Abstract
As new computing paradigm evolves in Information Technology, there is a necessity for the
methodologies and practices related to IT to also evolve.

The important aspect of Information Technology is the use of technology to enable business. Project
Management assists in managing the implementation of technology infrastructure for the enterprise
as projects by delivering business requirements in incremental steps, instead of a Big Bang
Approach.

Cloud Computing is evolved from the basic need for organizations to integrate and enable alignment
to changing business needs faster and seamlessly. It also provides the cost advantage by providing
infrastructure (IAAS), platform (PAAS) or software (SAAS) as a service. The availability of pervasive,
elastic, virtualized service with flexible costing, also possess a challenge for organizations to be able
manage change, resulting in increased risk of project failure due to delays, lower quality, increased
cost etc. There is increased Gold Plating challenge as Service providers sell add on services to
increase profits, or when project team delivers value beyond specified scope due to easy service
availability, without considering the challenges of integration, security and policies.

In addition to the fundamental project management principles, the new strategy for project
management for Cloud Computing will involve effective management of

Scoping and change management

Risks management

Cost management

Quality of Service for availability and reliability

In this paper, we focus on the various aspects of project management to mitigate risk of project failure
in a cloud computing paradigm through effective scoping, change and risk management.

2. Introduction
Projects often do not behave as expected; this is often the case with large complex projects. The
unexpectedness and the difficulties associated of the project increases while adopting new
technologies.

Although cloud computing is only a different way to distribute and consume computer resources, it is
still a new paradigm. There is a fundamental change in architecture, software, development
methodologies, and the way we store and consume information and more importantly project
management. The traditional project lifecycles will not be sufficient to ensure success of Cloud
computing projects. New skills need to be acquired by not only the project team but also the project
manager. There is a need to identify the early warnings more precisely. To manage the problems, we
need to first understand the problem areas.

4 Page

This paper will not discuss in depth about Cloud Computing concepts, but scan through the concepts
of cloud to identify the areas to watch out for early signs of failure. An in-depth analysis of the causes
of failure of the cloud computing projects reveals the possible solution to the issue. As project
managers it is important to address these solutions in the planning of the project.

3. What is Cloud Computing
Cloud Computing is a Network-based computing solution that delivers IT as a Service. The computers
in the cloud collaborate to collectively deliver computing resources. The allocation of the computing
resources is flexible and allocated based On-Demand. To organizations this paradigm shift helps in
reduced cost by reducing both capital expense and operational expenses. It allows to free valuable
personnel resources to focus on delivering value and at the same time is scalable to meet increasing
demands of a growing organization.

There are various layers in a cloud, but the three most important layers are IaaS, PaaS and SaaS.
Other layers include BPaaS (Business Process as a Service), CaaS (Communications as a Service)
etc.

Infrastructure Layer is the foundation, and consists of physical assets such as servers,
storage etc. In Infrastructure as a Service (IaaS), the control of the operating system, storage
and applications deployed is with the Organization.
Platform Layer provides access to the operating system and associated services. Platform as
a Service (PaaS) provides the organization the control on the applications that are deployed,
and to some degree the configurations of the hosting environment.
Application Layer most often referred as Software as a Service (SaaS) has applications that
are provided to users on demand. Organizations do not have any control on the hosting
environment.

There are different types of Cloud. A broad categorization includes Private Cloud, Public Cloud and
Hybrid Cloud.

Public Cloud: Resources are dynamically provisioned over Internet based applications, using
one or more data centers, shared among multiple customers with varying degrees of data
privacy and minimal organizational control on changes to applications.
Private Cloud: The computing architectures are similar to that of the Public Cloud, but the
services are delivered over the internet. The virtualization of computing resources helps
organizations make optimal use of the computing resources without the risk on compromising
the data privacy.
Hybrid Cloud: The combination of Public and Private Cloud to enable organizations have
larger control of the privacy and security aspects, while allowing access to scalable
computing resources of the public cloud that can be accessed On-Demand.

The advantages of adopting the cloud computing solutions are varied. The few advantages that are
easily recognized by organizations are

5 Page

Flexible Pricing Model: Free, Subscription model, Pay per use, Enterprise Unlimited etc.
Reduced IT spending: Capital expenses and Operational expenses Costs are reduced
Greater Flexibility, incredible scalability and Faster Implementations.
On Demand Service
Elimination of unnecessary hardware, more remote management
Leverage Technology Evolution

Common Causes of Failure in Cloud Computing Projects

90% of projects do not meet time/cost/quality targets. Only 9% of large, 16% of medium and 28% of
small company projects were completed on time, within budget and delivered measurable business
and stakeholder benefits (Nicoletti, 2012). There are many reasons for such failures.

Some of the common causes for project failures are:

Failure to define “Success”
Failure to quantify the expected advantages
Failure to hold yourself accountable
Failure to understand the Supply Chain
Failure to manage and monitor applications
Failure to understand the financial realities
No complete understanding of the requirements up front
Jumping too quickly into technology
Failure to understand the Legacy Data model and Flow
Unskilled Resources
Non Aligned Business and IT

A broad analysis of the causes shows that the issues are primarily related to Scope, Change, Risk
and Cost. A look at each of the causes in detail will reveal the solution to the problem.

Failure to define “Success”
Like traditional projects, cloud computing projects also require to be part of an enterprise wide
Strategy. Cloud Computing is not a solution to all the problems of an organization. Therefore it is
important to set a list of realistic expectations to be achieved as part of the Cloud Computing Shift.
For most organizations, the list is not known upfront. Cloud Implementations are generally adopted
with short term focus and in the long term these project fail to meet the high expectations of the
business users.

The solution to this concern is to establish a strategy and define a roadmap. If there is no strategies
established then start with a pilot and use the experience to define the strategy and create a roadmap
for the Cloud implementations.

6 Page

Failure to quantify the expected advantages
The stakeholders must be fully educated about the advantages and disadvantages of the Cloud.
Being aware of only the short term benefits gives the business users high expectations and failure to
meet the expectations can result in projects to be doomed failure.

It is also necessary to quantify the cost advantages, as a pay as you use model may not always be a
cost benefit in terms of Total Cost of Ownership (TCO). The pay as you use model provides cost
benefit when there is a need for faster and flexible scalability. The various factors influencing the TCO
of Cloud Computing include:

Not choosing a proper deployment model based on the needs of the applications
Planning for disasters, outage, failover, redundancy etc., as these can cause adverse effects
to business.
Long Term Use of Computing Resources (CAPEX+OPEX).
Planning for Migration strategies (Between Vendors or even between Public to Private
Clouds)

Failure to hold yourself accountable

The common misconception about cloud is that all the risk is with the Cloud Vendor. It is easy to
transfer the risk, but the organization itself is ultimately accountable for its reputation. Penalties
cannot pay for reputation; it can only pay for losses. It is therefore necessary to ensure that the
project team scrutinizes all security and availability aspects and create strategies and governance
processes.

The following issues need to be addressed as part of Cloud Implementations

Governance Processes
SLA for various services provided with penalty clauses
Availability / Up Time SLA
Plans for Failover, Redundancy, Backup and Disaster Recovery.

Failure to manage and monitor applications
Another common mistake made by organizations migrating to the cloud is to believe that the
management and performance problems will not exist. The project team needs to understand that the
responsibility of ensuring the availability and meeting the performance requirements of the
applications is their responsibility. With the adoption of cloud this responsibility diminishes
considerably, but does not vanish.

A successful project needs to incorporate these aspects into the planning of the project, time and
effort needs to be invested to ensure that these criteria are met. Failure to address this concern
upfront will result in unsatisfied users and finally a failed project.

Failure to understand the financial realities
One of the primary reasons for an organization adopts cloud is the cost advantage. But the financial
aspect depends on the type of service consumed, who pays, how the payment is made.

7 Page

It is critical for the project team to do a financial analysis on the TCO to identify the actual cost vs. the
perceived cost benefit.

Most of the projects today are never completed within Budget. This becomes even more true to Cloud
Projects as features is available as Service, and the stakeholders and project team can be prompted
to utilize a service due to its lower initial acquisition cost. It is therefore very important to continuously
monitor scope creep and have effective change control measures.

No complete understanding of the requirements
When projects are initiated without a complete understanding of the requirements, the chances of
failure are high. In a hurry to adopt cloud, and to reduce cost the project team tends to move quickly
to implementation without a roadmap to guide the implementations. As discussed earlier, it is
important to have a roadmap. But at the same time it is important for the project team to realize that
the organizational dynamics change over a period of time.

With Cloud, it will not be beneficial to attempt long term projects. Instead, the Project management
strategy should be agile in nature. This methodology provides the project team many opportunities to
assess the project and respond to change requests faster. Also, agile methodology provides the
project team to incrementally deliver value. The methodology allows for well-defined set of changes
and added features to be addressed in sequential phases. This helps reduce development cost and
time to market (TTM) and users’ needs are addressed quickly. Scope Creep and Gold Plating is
controlled as new requirements or change requests are address in successive iterations and after
complete analysis of the requirements. Also the iterative approach enables the project team and other
stakeholders to associate a perceived cost (if not an actual cost) to the requirements. The iterative
approach also enables the committed features to be delivered timely, thus reducing the cost
escalations.

Jumping too quickly into technology
Deciding the technology before having a finite idea of the requirements can prove to be deadly in
cloud projects. The few problems are outlined below

Since cloud is a service based implementation, it would be costly to change vendors half way
through the implementations
Migrating applications will be difficult as there is little or no standardization between different
implementations and vendors.
A semi Optimal Solution will hinder the organization goals and also the long term strategy as
users will be reluctant to adopt any further implementations.

Thus it is important to scrutinize the organizational objectives and roadmap, and understand the user
requirements before the technology selection is initiated. The technology selection should also be
based on the security, availability, architecture and long term performance considerations.

Failure to understand the Legacy Data model and Flow
In most Organizations the cloud migration team is separate from the project team that is involved in
development and maintenance of the existing applications. The reasons for this are primarily:

The need for a different resource skill set to accomplish cloud projects

8 Page

The need for organizations to keep the business running while adapting to the cloud
Only sections of the applications are migrated depending on the compliance requirements of
the applications

The above reasons mandate that the current applications and the data model are well understood by
the Cloud migration team, as it is necessary to integrate the applications. Also, cloud is a new
paradigm. It is necessary to make the applications lean by eliminating the unwanted processes. Re-
engineering eliminates waste and conserves resources. It is possible to digitize the processes by
making best use of the available technologies. This also helps in adopting the iterative methodology
to provide continuous improvement.

If the cloud migration team fails to understand the legacy data model and flow, the users will not get
the functionality that is intended and the migration to cloud is bound to fail.

Unskilled Resources
Cloud computing resources are required to possess a different technical capability and skills for
integrating the On-Premise applications and the Cloud based applications. As cloud is a relatively
new technological paradigm there is scarcity of skilled resources. The risk of project failure becomes
drastically high with the lack of required skill set. The mitigation plans to address this risk is:

Providing the right amount of training to the resources involved.
Leveraging consultants and outside experts to address the technology challenges.

This would increase the overall cost of cloud implementations, but would be much less than a failure.
The important activity is to plan for this expense upfront and mitigate the risk earlier in the project
initiation phase and make a smooth progress to the finish line.

Non Aligned Business and IT
Cloud meets the time to market demands of business, without the need to wait for IT to deliver
solutions. Due to the ease of service availability, the business is adopting cloud at a much faster rate
than IT. This non alignment can cause the organizational entropy to increase. There would be
redundancy in process, data and applications and over a period of time, the organizations would find
it difficult to maintain coherence. In the long term, as entropy increases, enterprises will find it difficult
to realign IT and cloud solutions to meet Organizational strategies. This is against the basic reason of
adopting cloud – Faster time to market.

Also, Business may not be able to differentiate the nonfunctional aspects of a cloud solution, which IT
can easily analyze. A cloud project will not be successful until the organizations interest is aligned and
all stakeholder groups commit to its success.

Failure to understand the Supply Chain
Supply Chain is an increasingly important aspect in Cloud Computing Projects. The project Manager
has an added responsibility to ensure that the Cloud Service Provider will be able to address the
project requirements over the long term. The following aspects need to be validated while choosing
the Cloud Service Provider

Financial Stability to ensure ability to provide continued service

9 Page

Implementation Expertise in similar projects
SLA and track record
Availability Conditions and clauses

It is suggested to go with reputed service providers, as their expertise will be an added advantage in
project risk mitigation strategy development.

4. Conclusion

Based on literature from IBM (IBM Academy of Technology, 2010), the basis of cloud implementation
is a set of well-defined, proven process as illustrated in Figure 1. Service management is vital to
being able to deliver, or even access Cloud enabled services.

Figure 1: Cloud service management implementation framework
The framework shows that people, process, tools and information needs to interact to enable a
successful cloud implementation. It is also observed that some of the causes of project failures are
inter-related. Failure to address one can result in the other. E.g. failure to understand the financial
realities can increase TCO which in turn can impact the expectation of reduced cost. Or, not
understanding the requirements completely will result in inadequate technology selection.

Analysis of the causes of failures clearly indicates that it is important to have a clearly defined
strategy and well defined roadmap to enable success. It is also noted that managing requirements,
scope and changes are an important aspect to ensure success of cloud computing projects.

Identifying risks and developing mitigation strategies early in the project can boost the success rate.
Also, it is necessary to watch out for early warning signs to ensure that the problems are addressed
early.

Agile or Iterative methodology of project management is the most suitable in cloud projects as it
allows attaining the objective of reduced TTM while allowing addressing the challenges of changing
business needs. It also helps control scope creep and gold plating through effective change control by
addressing changes as part of separate iterations.

10 Page

5. References
[1] Agile Methodology, "Understanding Agile Methodology", http://agilemethodology.org.
[2] Engira Ram, "Cloud computing: Technology, Risks, and Controls", http://www.nyiia.org, April 2010.
[3] Healing Eric, "Cloud Computing Risk Management", Deloitte Development LLC, Presentation,
October 2011.
[4] IBM Academy of Technology, "Cloud computing insights from110 implementation projects", IBM
Academy of Technology Survey, IBM, October 2010, pp. 1-12.
[5] Nicoletti Bernardo, "Project Management and Cloud computing", PM World Today,
http://www.pmworldtoday.net, January 2012 (Vol. XIV, Issue I), pp. 1-11.
[6] Walker Grace, "Cloud computing fundamentals", IBM Developer Works, IBM, December 2010, pp.
1-10.

6. Author’s Profile:
Praveena Parameswaran is Technology Architect and associated
with Infosys Technologies, India. She has 14 years of IT
experience in data warehousing, enterprise architecture, project
management. She has received a She is B.Tech in Electrical and
Electronics Engineering from University of Calicut, India in 1997
and a MS in Microelectronics from Manipal University, India. She
is Zachman Certified Enterprise Architect Associate and has
managed multiple projects to align IT and business strategies. Her
area of interest is Cloud Computing and Big Data

Email: pravina.pv@gmail.com

11 Page

ETCA_5

More Related Content

What's hot (19)

Similar to ETCA_5 (20)

More from PMI2011 (20)

Recently uploaded (20)

ETCA_5