White Paper

A Solution Blueprint for DevOps

Executive summary This white paper prescribes a DevOps solution blueprint that incorporates the following
capabilities:
Industry surveys and reports describe the
highest achieving companies as those turning ƒƒ Integrated architecture for optimum CI, CT, CD and CCM operations
to a DevOps infrastructure implemented in ƒƒ Hierarchical test execution model to automate all lifecycle test phases
accordance with best practices to support
ƒƒ Tools framework including Restful APIs to orchestrate all DevOps operations
rapid-paced Agile and non-Agile based product
development processes with ever more ƒƒ Expert services that support a logical phased implementation plan and a smooth

frequent release deployments to customers. transition from the existing infrastructure minimizing disturbances of the existing

But many organizations are challenged by the development, test and release process

migration to these new processes. ƒƒ Key Performance Indicators (KPIs) and a dashboard that provides the requisite
visible measures necessary for Continuous Change Management (CCM) and
The DevOps solution blueprint encompasses
monitor operational performance of the DevOps system itself to guide operations
Continuous Integration (CI), Continuous
and continuous improvements.
Test (CT), Continuous Delivery (CD) and
Continuous Change Management (CCM) Without continuous control, the code development complexity curve becomes unstable,

capabilities with automated orchestration of creating a “big bang process” that results in process failures such as late deliveries,

all operations necessary for the rapid-paced poor quality release content and cost overruns. CT together with CCM delivers progress

product development essential for business monitoring and control features required to reduce complexity by ensuring corrective

performance. actions are taken early and often before problems build up into a big bang.

The DevOps solution blueprint applies whether an organization is just getting started
with Agile or currently employs Agile, waterfall or waterfall-scrum processes. It also
applies whether the environment is hosted on physical systems, virtual systems or a
combination of the two.

Readers!

Please take note: acronyms, abbreviations and definitions

used in this document are listed at the end of this document.
White Paper

A Solution Blueprint for DevOps

While the implementation of a best practices DevOps Problem and historical context
infrastructure is not easy, the benefits are compelling:
Problems context
ƒƒ Catch product defects as early as possible in the
development cycle Software development best practices are trending toward
ever more rapid and frequent release deployments to
ƒƒ Perform CI, CT and CD in frequent and short cycles
customers. The Agile software development methodology
ƒƒ CapEx and OpEx expenses saved due to efficient is typically employed. An effective Agile environment
resource monitoring dictates that software development work be done in small
ƒƒ Accelerates product delivery velocity, innovation and steps within continuous cycles and that new software
time-to-market feature and fix releases be fast-and-frequent-to-market
so customers get new features and fixes quickly, and
ƒƒ Improves infrastructure administration
suppliers get revenue and customer feedback quickly.
ƒƒ Improves process controls with data-based dashboard The efficiency of releases is a competitive advantage for a
decision tools business because it can have a major impact on business
ƒƒ Reduces down time by ensuring effective backup, performance.
archival and recovery
In environments where best practices are invoked, the
ƒƒ Improves Intellectual property protection and security Agile software development process is supported by a
The implementation of a best practices DevOps solution highly automated and controlled DevOps infrastructure.
requires an array of capabilities including test orchestration At the heart of a DevOps system, which is implemented
tools, assessment services, 3rd party integration services, according to best practices, are efficient continuous
KPI/dashboards, and resource utilization optimizations that operations that automatically orchestrate all the tools
may include virtualization of DevOps components within and process components needed to realize a rapid
private and hybrid cloud deployments. Case studies based cycle including source code changes, merging of code
on real deployments by experienced DevOps experts changes, host system initialization, build process execution,
have proven that a DevOps solution, when implemented automated release packaging, multiple levels of automated
according to best practices, quickly delivers powerful and testing, and automated results reporting and analysis. The
compelling ROI. end-to-end integration cycle, from code change commit,
to reporting of results, must be extremely fast, usually in
the order of minutes or hours compared to older software
development practices that typically take days to complete
the equivalent steps. The speed and frequency of the
DevOps cycles necessitates automated orchestration of all
CI, CT, CD and CCM operations.

2 | spirent.com
When the DevOps infrastructure is not implemented In a DevOps environment implemented in accordance
according to best practices, then the following problems with best practices, elastic on-demand build and test
typically result: resources accomplish fast, frequent cycles, without
sacrificing test coverage or dedicating costly under-
ƒƒ Large, complex software merges bottle-neck
utilized equipment. DevOps environments may demand
integration because large merges require longer
multiple copies of systems which is expensive if those
build and test times which violate the best practice
resources are dedicated physical systems. The solution
that prescribes frequent, quick change cycles.
to control the cost of multiple on-demand computing
ƒƒ Excessively long software build and test cycles resources needed for parallel building and testing is to
caused by lack adequate resources in the build and virtualize the components of the system being built and
test environment. tested and also virtualize the test systems whenever
ƒƒ Proliferation of costly dedicated equipment to ensure possible. In this way a test system/system under test
there are sufficient resources to meet peak demands combination can be deployed on-demand as virtualized
of the DevOps system. resources. The use of virtualized resources allows the
host equipment to be deployed on-demand and re-used
ƒƒ High rate of defects found by Quality Assurance (QA)
for other code lines or other purposes when not needed
or customers causes by rapid but inadequate test
for a particular product or code line build or test job.
coverage.

ƒƒ Time-consuming defect troubleshooting and fixes if

there are too many changes or failed tests per cycle.

ƒƒ Integration problems found late in the release

may occur if the test coverage is not continuously
checked during the development period.

ƒƒ Missed delivery dates due to efficiency and quality

problems.

ƒƒ Inadequate process control in which problems persist

across multiple code commit cycles without being
noticed until later, necessitating costly reverts to
earlier versions in order to fix the cumulative effects
of the problems which crept in over time.

ƒƒ Fault -intolerance, backup problems, and recovery

problems are very costly and prevalent in poorly
designed DevOps systems that fail to consider the
need to have redundancy, and frequent back-ups for
all the build, test, and tool data.

ƒƒ Security and Intellectual Property protection

problems occur when access restrictions to ALL of
the components are overlooked. This may occur
when the various DevOps system components do
not follow a consistent security model for access
controls and don’t take care to obfuscate credentials
or source code IP within orchestration scripts.

spirent.com | 3
White Paper

A Solution Blueprint for DevOps

Requirement
Design
Design
Analyze
SW Change Software
Change
Merge
Build Test Merge
Package
Test
Build
Analyze
Package
Release

Waterfall Model Agile Model

Historical context
Figure 1: Waterfall Prior to Agile, the most prevalent software development According to http://agilemethodology.org/ “It’s easy to
and Agile Software methodology was the sequential “Waterfall model”. see the problems with the waterfall method. It assumes
Development Models
Figure 1 compares the sequential nature of waterfall with that every requirement can be identified before any
the iterative, cyclic nature of Agile. Agile process typically design or coding occurs. Could you tell a team of
would iterate many times to achieve the same results as developers everything that needed to be in a software
one waterfall cycle, but with much less effort and higher product before any of it was up and running? Or would it
quality. be easier to describe your vision to the team if you could
react to functional software? Many software developers
In the waterfall model it is assumed each step in the
have learned the answer to that question the hard way: At
process is completed for the entire system being
the end of a project, a team might have built the software
developed prior to completing the next step. This
it was asked to build, but, in the time it took to create,
resulted in a long time between requirement definition
business realities have changed so dramatically that the
and release to customers. Each step is dependent
product is irrelevant. Your company has spent time and
on the prior step so problems in any one step would
money to create software that no one wants. Couldn’t it
cause delays for the entire project. All too often the
have been possible to ensure the end product would still
long development period and the stepwise delays
be relevant before it was actually finished?”
causes projects to miss their deadlines. In many cases,
the customer requirements change by the time the
project is completed. The problems associated with the
Waterfall Model ultimately drove the creation of the Agile
Development Model.

4 | spirent.com
 DevOps Context
Engineering / QA DevOps IT Operations

Continuous Configuration

Automated
CI CT CD
Continuous Testing
Continuous Integration Continuous Testing Continuous Delivery

Continuous Control

Agile Infrastructure
Shorter release cycles, predictable dates & higher quality
CI = Continuous Integration, CT = Continuous Testing,
CD = Continuous Delivery, CCM = Continuous Change Management

According to http://en.wikipedia.org/wiki/Agile_ Agile software development involves continuous Figure 2: Continuous

software_development “Agile software development integration (CI) cycles and each CI employs automated Testing is the heart of the
is a group of software development methods in continuous testing (CT) to verify the result before the next DevOps System
which requirements and solutions evolve through change cycle is started. In this way the software changes
collaboration between self-organizing, cross-functional are controlled continuously.
teams. It promotes adaptive planning, evolutionary
An efficient DevOps infrastructure consisting of
development, early delivery, continuous improvement
automated processes, tools and methodologies is
and encourages rapid and flexible response to change.
essential to fully realize the goals of Agile and rapid
It is a conceptual framework that focuses on delivering
product development. According to http://en.wikipedia.
working software with the minimum amount of work.
org/wiki/DevOps, DevOps “aims to help an organization
The Agile Manifesto introduced the term in 2001. Since
rapidly produce software products and services. Simple
then, the Agile Movement, with all its values, principles,
processes become clearly articulated using a DevOps
methods, practices, tools, champions and practitioners,
approach. The goal is to maximize the predictability,
philosophies and cultures, has significantly changed
efficiency, security and maintainability of operational
the landscape of the modern software engineering
processes. This objective is very often supported by
and commercial software development in the Internet
automation.”
era.” Figure 1 shows the relevant concepts of the Agile
Software Development Model. Figure 2 illustrates the DevOps Model and shows how
Continuous Testing is at the heart of system.

spirent.com | 5
White Paper

A Solution Blueprint for DevOps

DevOPs solution blueprint architecture

The DEVOPS Solution Blueprint architecture, presented The DevOps Solution Blueprint architecture is composed
in Figure 3, addresses the problems described in of the following modules:
section 2.1 by implementing Continuous Integration
1. Dashboard: At the highest level of the architecture
(CI), Continuous Testing (CT), Continuous Deployment
a dashboard provides information for continuous
(CD) and Continuous Change Management (CCM) sub-
change management. The actual dashboard
systems within a modern DevOps infrastructure context
implementation varies according the Key
described in section 2.2. The DevOps solution blueprint
Performance Indicators (KPIs) that are most relevant
architecture addresses these problems regardless of
to a particular customer. Therefore the Dashboard is
whether they occur in Agile or non-Agile development
typically customized for each installation.
environments. The DevOps blueprint supports physical,
virtualized, and mixed physical/virtualized host and 2. Continuous Integration: Automation framework
lab network environments. The DevOps blueprint that provides centralized orchestration of the entire
architecture supports configuration and control across DevOps system.
multiple product boundaries, service layers, and software 3. Source and Artifact Control: Database of source
versions, even in cases where systems must be verified in code. Software packages that are used by CI system
parallel, on-demand, and in real time. build processes.

4. Test Management: Database of test scripts and

CI / CT dashboard artifacts needed for testing

5. Orchestration Automation: automates tests and test

Continuous Source & Test
integration artifact control management topologies

6. Build and Test Infrastructure: Virtual and/or physical

Orchestration automation test beds.

(Build and Test) infrastructure

Physical
Virtualizaton
testbed
Figure 3: DevOps
Solution Blueprint
Architecture

6 | spirent.com
 Development
Continuous Change Management (CCM)

SVM
Source Code ready
P4, Git, SVN, etc.
not ready
Analytics
Dashboards

Commit
trigger checkout
push / pull
Continuous Integration (CI)

Continuous Delivery (CD)

results

CI / CD Artifact repository Deliver

Orchestration Tool Images, tests, deliverables, logs, results

commands
and images images logs
responses

Build(s) SUT
Dev, CI, QA, Labs

test info test response info

Test I/P
Test and lab Test O/P
management
Virtual and / or physical environments
Continuous Test (CT)

DevOps tools framework

The DevOps Solution Blueprint framework presented in 2. A CI/CD Orchestration Tool is a programmable Figure 4: DevOps
Figure 4 consists of the following elements: system that executes scheduled and event-driven, Solution Blueprint Tools
Framework
jobs. The most popular CI/CD orchestration tool is
1. A Software Version Management (SVM) system
Jenkins.
supports software development by providing
software configuration control and identification, 3. A Build System compiles and packages build
visibility into the status of software changes, and artifacts. CI provides rapid feedback to the entire
management of versions of the tools needed to team about the current state of the software. This
implement the build process. In a best practices feedback improves quality, lowers costs, and
DevOps system the SVM includes a well-designed, enables management to make informed decisions
flexible repository structure and supports effective on how to most effectively allocate resources to
software branching and merging capabilities. This meet mission objectives. DevOps leverages this
capability is critically important for Agile development practice by automating the build process and then
processes, as developers merge changes back into triggering the start of a quality cycle when a set of
the repository frequently. Effective Software Change conditions is met. A trigger is typically a commit
Management (SVM) systems often include a mix of into a monitored branch, though multiple commits
open-source and commercial software with small, can be cached and generate a single trigger event.
custom modules tailored to the specific development DevOps orchestration tools manage the build
organization. Examples of SVM Systems are: servers, physical and/or virtual, spinning them up
Subversion, Git, Mercurial, Perforce, and ClearCase. as needed. This feature allows the organization to
efficiently make use of private, hybrid or public cloud
computing resources.

spirent.com | 7
White Paper

A Solution Blueprint for DevOps

Commit

Notify &
Trigger
Report

Destroy CDBuild
DUTs

Create
L1 Test
DUTs

Deploy

Figure 5: DevOps Cycle 4. An Artifacts Repository contains completed build 6. Test and Lab Management Systems may also be
with Virtualized SUT packages and information required to test the hosted on physical or virtualized platforms. In a
packaged builds. The repository is an organized DevOps system implemented according to best
store of artifacts related to the product. In addition practices, functional and system level QA testing are
to source and build artifacts the repository may fully automated. Many organizations are reluctant
include, test artifacts such as test scripts, logs and to perform automated QA testing as part of CI due
test results data related to issue and ticket tracking, to increased cycle times and unreliable automation.
compliance and regulatory artifacts, development A DevOps system implemented according to best
and user documentation, and configuration data. practices, is able execute stable QA regression runs
Examples of Artifacts Repositories include an NFS while dynamically managing virtualized and physical
file share, Zephyr and HPQC. The repository may hardware resources and scale the tests dynamically
be implemented as a single database or multiple according to resource availability for the allotted
databases. test time. Examples of Test and Lab Management
Orchestrations Systems are Spirent’s iTest and
5. The software for the System Under Test (SUT) or
Velocity.
device under test (DUT) may be hosted by a physical
or virtualized device, or network of devices. As 7. A Delivery system packages release artifacts such
shown in Figure 5, when the SUT is virtualized the as software packages in formats consumable by
CI cycle includes steps to create virtual machines customers and provides controls to determine when
to host the SUT software and when the tests are to promote a packaged product version to customer
completed the DevOps system destroys the SUTs so release status.
the host resources are available for other purposes.

8 | spirent.com
8. Analytics collect and analyze reports and logs from The increasing number of organizations adopting the
the CI, CT and CD systems to determine whether cloud model indicates the value they are finding in
any additional changes are required to complete a the cloud. Cloud computing is an extremely cost-
quality cycle. For example, tests that result in failures effective method of deploying, maintaining and
may automatically determine that a change in the securing desktop and server applications. Rather
SVM system must be reverted and may automatically than purchase hardware and software licenses for
trigger the revert process. each member, an organization can serve them up
based on demand, not only reducing costs but also
9. A DevOps host environment may be private, public
enhancing scalability. This applies to DevOps CI, CT,
or hybrid cloud services (e.g. AWS, Rackspace).
CD and CCM services also.
The cloud is an abstract term used to generalize
various types of remote storage, computation By virtualizing DevOps applications, an organization

and user access services. While cloud-computing can centralize and streamline management

infrastructures may consist of many types of services, processes. Tasks such as backup, recovery, and

most implementations provide one or more of the installing updates and security patches can be

following service models: performed quickly. Cloud computing relieves

IT departments of a great deal of the logistical
a. Infrastructure as a Service (IaaS) – This service burden of maintaining a desktop environment.
model provides basic computing functionality, Installs, updates, and other maintenance items are
typically in the form of a Virtual Desktop accomplished at a central location rather than having
Infrastructure consisting of virtual machines and to travel to the user’s site.
support resources. The users typically access their
The Virtual Machine environment (e.g. VMware
desktops from a software client running on low-
VSphere) may be a variety of hypervisors and
cost terminal hardware.
containers depending on the specific environments
b. Platform as a Service (PaaS) – This service model that each tool supports. The DevOps blueprint is
provides platform specific services, typically in agnostic regarding specific hardware or cloud-based
the form of virtualized servers such as Software virtualization technologies. Delivering build, test,
Defined Networking (SDN), domain controllers, and assessment services on these platforms allows
web, database, and development servers. the CI, CT, CD and CCM process to be extremely
stable and elastic. The ability to perform DevOps at
c. Software as a Service (SaaS) – This service model scale in a private or public cloud is a key driver for
provides access to virtualized software application reducing CapEx and OpEx costs associated with
running in the cloud. This feature provides software development. In a fully virtualized DevOps
platform independence and does not require the environment both the test system and SUTs use VMs
user to install or maintain the application locally. that are deployed on-demand.

10. The DevOps blueprint solution includes back-up,

and recovery of build and test artifacts such that
in the case of system failure the entire system can
be recovered from archived or backed up artifacts
quickly.

11. The DevOps blueprint solution obfuscates system

credentials and provides access controls to ensure
unauthorized access to sensitive source code and
Intellectual Property. 

spirent.com | 9
White Paper

A Solution Blueprint for DevOps

Hierarchical DevOps solution blueprint test execution model

To fully integrate QA testing, a hierarchical model for test The levels listed above provide different types of QA
execution is recommended. This model consists of three coverage and are executed at different intervals. L1
levels of testing: testing is a breadth-first test cycle focused on testing
the entire system with each quality cycle. L1 regression
ƒƒ Test Level 1 (L1)–QA regression run as part of CI cycle
runs are kept short and typically include strategic test
‚‚ Software static analysis points across the system. These test points focus on
• Compiler warnings the integration of major components and detecting
integration defects. L2 and L3 test are typically scheduled
• Coding standards compliance (Format and
rather than being run as part of CI. The L2 regression is
metrics)
typically run each night and includes tests focused on
• Security assessment (Unsafe coding practices, system components that are changing in the current
implementation vulnerability detection) release. The L3 regression is a full battery of tests
‚‚ Software dynamic analysis focused on detecting low-level defects. L3 is typically
scheduled on a weekly basis or as part of release
• Memory re-use and/or leaks
acceptance.
• Thread contention/deadlock, multi-core scaling,
cache access

• Undefined behavior

‚‚ Unit test and code coverage

ƒƒ Test Level 2 (L2)–Nightly Integration and regression

‚‚ End-to-end feature testing

‚‚ Hardware/software integration test

ƒƒ Test Level 3 (L3)–Weekly or release

• Interoperability compliance (3GPP, IETF, ITU)

• Regulatory compliance (FCC, HIPAA, FISMA)

• Product/system development security

assessment (HIPAA, PCI, NIST)

• Performance assessment (Bandwidth, error rate,

signal to noise ratio)

• Operational security assessment (HIPAA, PCI,

NIST)

10 | spirent.com
 Capability • Analyze existing build, test and deployment environment
assessment • Assess the existing virtualization, CI and automation capabilities

• Implement new CI capability

Virtualization • Enhance existing virtualization capability

Documentation
Continuous • Implement new CI capability
integration • Enhance existing CI capability

Training
Continuous • Integrate testing into the Continuous Quality cycle
testing • Automate QA testing and determine automation platform

Automate • Test automated VM control, build, test and deployment workflows

orchestration • Update source control and artifact management

Dashboard and • Realization based on customer environment and

data visualization organizational processes

DevOps solution blueprint services KPI and Dashboards

No two organizations are the same. Many organizations Metrics or Key Performance Indicators (KPI) are frequently Figure 6: DevOps
have some level of automation, CI, test and virtualization generated and updated to provide timely status Blueprint Solution
Services
but may not be operating at best practices levels for all reporting to organizational leadership. The specific KPI
parts of their DevOps CI, CT, CD and CCM sub-systems. and presentation will vary depending on organization
It is important to engage experienced experts to assess preferences.
an organization’s current DevOps status, goals and
The following are examples of measureable KPI’s that are
implementation plans. Figure 6 defines DevOps Solution
relevant and can be derived from DevOps deployments:
Blueprint Services that need to be considered.

ƒƒ Percent of tests automated

Through the capability assessment, a strategic plan
is developed that will migrate the team’s existing ƒƒ Percent availability of automation test infrastructure
infrastructure, with minimal impact to current operations. ƒƒ Execution times for builds and tests

ƒƒ Product Velocity: release frequency

ƒƒ Test efficiency (e.g. number of acknowledged

bugs divided by the total number of potential bugs
reported)

ƒƒ Total Capex and Opex cost reduction compared to

other approaches (e.g. ROI case study)

spirent.com | 11
White Paper

A Solution Blueprint for DevOps

Benefits of the Devops solution blueprint

Figure 7 illustrates the primary benefit of a DevOps The DevOps Solution Blueprint benefits include:
system implemented according to a best practices
ƒƒ Catch issues as early as possible
solution blueprint:
‚‚ Perform CI (build) and CT (test) in short, frequent
To provide automated continuous control of the entire
cycles
development-test-release cycle in the context of Agile
ƒƒ Save on CapEx
and non-Agile rapid continuous integration complexity.
‚‚ Achieve efficient utilization of virtualization
Without continuous control, the complexity curve
environment in combination with physical test beds
becomes unstable, creating a “big-bang” process that
ƒƒ Save on OpEx
results in failures such as late deliveries and poor quality
release content. CT and CCM bring the control required ‚‚ Create control sequences quickly. Use
to reduce complexity. Orchestration Automation to enforce change
management and consistency in reporting, build
environments, and test-bed interactions. Achieve
a consistent, reliable, repeatable development
process with a consistently managed environment.
Complexity

Big Bang ƒƒ Focus on your business and accelerate time to

market

Reduced ‚‚ Follow best practices delivered in professional

complexity services and tooling to focus on your products and
not the supporting infrastructure.

ƒƒ Make evidence-based decisions

Development time ‚‚ Promote data-based decision by keeping

Figure 7: DevOps everyone on the same page all the time with a CI/
controls development CT dashboard customized around your internal
time despite increasing
processes and existing systems.
complexity

12 | spirent.com
Summary
Whether an organization is just getting started with Agile
or similar fast paced product development, or currently
employs Agile or DevOps processes but wants to
improve the efficiency of their environment the DevOps
solution blueprint described in this document provides
a comprehensive framework and a step-wise approach
towards implementing a solution.

Anyone considering installing or updating their

Continuous Integration environment can utilize the
DevOps Solution Blueprint to accomplish:

ƒƒ Solution architecture suitable for optimum

Continuous Integration

ƒƒ Hierarchical Test Execution Model

ƒƒ Tools framework including orchestration of CI, CT, CD

cycles

ƒƒ Virtualized CI, CT setups

ƒƒ Services that support the strategic improvement

plan

ƒƒ A logical phased implementation plan that minimizes

disturbances of the current system

ƒƒ KPIs and Dashboard to measure performance of the

CCM solution

spirent.com | 13
White Paper

A Solution Blueprint for DevOps

Spirent’s capabilities for DevOps

Spirent has expertise and experience implementing To support virtualized and cloud-based DevOps
successful DevOps deployments consistent with best deployments Spirent has developed plug-ins.
practices and in accordance with the DevOps Solution Additionally, Spirent has native session types that allow
Blueprint described in this white paper. direct communications with enterprise virtualization
management platforms.
Spirent, through its test tool products and professional
services capabilities, brings together open source Spirent professional services support DevOps upgrades
software, third-party commercial software, and unique or entire DevOps implementations depending on the
test automation capabilities to enable customers to customer goals and assessments. Project management,
improve quality and productivity in a scalable and cost- integration of iTest orchestration, implementation of
effective DevOps solution called “CLEAR DevOps”. virtualization and DevOps dashboards are examples of
project components that Spirent offers.
The Spirent iTest and Velocity products are perfectly
suited to provide test orchestration for best practices Figure 8 presents an overview of the phased approach
DevOps deployments . At the heart of the CLEAR Spirent developed for CLEAR DevOps.
DevOps solution is an integration of Spirent CT tools with
Case studies are available that demonstrate the benefits
CI called the Efficient Virtualized Continuous Integration
accomplished by Spirent implemented Clear DevOps
(EVCI).
(powered by EVCI).

Phase I Phase II Phase III

Reference implementation Cleanroom implementation Operational system

• Customer product agnostic • No impact to production • Transition
• Simple EVCI instance • Customer product under CI • Monitor
• Assess current capabilities • Customize • Advanced training
• Customer evaluates EVCI • Stabilize • Continued support
• Scale
• Basic training

Figure 8: CLEAR
DevOps Phased 1-2 weeks 4-8 weeks 2+ months
Implementation
Approach

14 | spirent.com
Reference links Acronyms, abbreviations, and
Agile: software development methodology: http://www. definitions
agilealliance.org/the-alliance/the-agile-manifesto/ API Application Programming Interface
BDD Behavior Driven Design
DevOps: software development method: http://
en.wikipedia.org/wiki/DevOps CapEx Capital Expense
CD Continuous Delivery
Jenkins: An extendable open source continuous CI Continuous Integration
integration server: http://jenkins-ci.org
CLEAR Collaborative, Leadership, Exploration,

Jira: proprietary issue tracking product, developed by Acceleration, Realization

Atlassian. It provides bug tracking, issue tracking, and CT Continuous Test

project management functions https://www.atlassian. DevOps Development Operations
com/software/jira DUT Device Under Test
EC2 Amazon Cloud Service
Open source software for building private and public
EVCI Efficient Virtualized Continuous Integration
clouds: http://www.openstack.org/
IaaS Infrastructure as a Service
Spirent iTest: http://www.spirent.com/Products/iTest ILO iTest Lab Optimizer
iTest Spirent Capture Execute Report test tool
VMware: http://www.vmware.com/
ITO Infrastructure Test Optimization
Zephyr: Real Time Test Management System: http:// Jenkins extendable open source continuous i
www.getzephyr.com/ ntegration server
Jira issue tracking product
KPI Key Performance Indicator
KVM Hypervisor
OpEx Operating Expense
PaaS Platform as a Service
QA Quality Assurance
QEMU Hypervisor
SCM Software Change Management
STC Spirent Test Center
SUT System Under Test
SVN Subversion
TaaS Testing as a Service
TDD Test Driven Design
VM Virtual Machine
VMware Company which produces VSphere
VSphere Server virtualization platform
Xen Hypervisor

spirent.com | 15
White Paper

A Solution Blueprint for DevOps

spirent.com
AMERICAS 1-800-SPIRENT
+1-818-676-2683 | sales@spirent.com
© 2015 Spirent. All Rights Reserved.
EUROPE AND THE MIDDLE EAST All of the company names and/or brand names and/or product names referred to in this document, in particular, the name “Spirent”
+44 (0) 1293 767979 | emeainfo@spirent.com and its logo device, are either registered trademarks or trademarks of Spirent plc and its subsidiaries, pending registration in
accordance with relevant national laws. All other registered trademarks or trademarks are the property of their respective owners.
ASIA AND THE PACIFIC
+86-10-8518-2539 | salesasia@spirent.com The information contained in this document is subject to change without notice and does not represent a commitment on the part
of Spirent. The information in this document is believed to be accurate and reliable; however, Spirent assumes no responsibility or
liability for any errors or inaccuracies that may appear in the document. Rev A. 01/15