INTRODUCTION - University of Albertarsawh/VMWARE-EMC-DELL-DEMO... · 2008-04-01 · Accelerating...
Transcript of INTRODUCTION - University of Albertarsawh/VMWARE-EMC-DELL-DEMO... · 2008-04-01 · Accelerating...
Vir tualization provides opportunities to reduce
complexity, improve service levels to the business
and lower capital and operating costs to provide
and maintain IT infrastructure.
Based on the experiences of our customers,
implementing virtual infrastructure is completely
doable and manageable. As a mature technol-
ogy, vir tualization can touch on a broad set of IT
stakeholders and IT processes. Cultural resistance
can stall or l imit many deployments, par ticularly
in larger enterprise organizations. Thus we recom-
mend that IT executives proactively define the
organizational vision for vir tual infrastruc-
ture, high-level goals and objectives to hone
vir tualization technical sk il ls, and vir tual infrastruc-
ture operational processes that can maximize and
accelerate the benefits of vir tual infrastructure.
I N T R O D U C T I O N
I N T R O
2
V I R T UA L I N F R A S T R U C T U R E O P E R AT I O N S
“ T h e o rg a n i z a t i o n a l m i n d s h a r e , ca p a b i l i t i e s a n d s k i l l s e t s t o a s s e s s , d e s i g n , d e p l o y, m a n a g e a n d s ca l e v i r t u a l i n f ra s t r u c t u r e .”
VI M indshare
VI Dr ivers
V I Sk i l l Sets
Establishing and maintaining a Vision for VI
Assessing Business Drivers for VI
Measuring Vir tual Maturity / Center of Excellence
Online Executive Workbook
Portal and Wik i
Whitepaper
This Workbook is organized in 4 sections: Vision, Strategy, Key Consid-
erations and Executive Passdown. Each section is designed to allow
you to capture new ideas and recommendation as well as to develop
your existing ideas and plans regarding your organization’s use and
implementation of virtual infrastructure.
YO U R E X E C U T I V E W O R K B O O K
D E F I N I T I O N
OT h E R TO O L S A N D R E S O U R C E S
V I S I O NS TR AT E G Y
K E Y CO N S I D E R AT I O N SPA S S D O W N
To achieve the benefits of virtualization beyond tactical and isolated
project oriented deployment, many cultural and organizational chal-
lenges that may stall or delay the process must be overcome. Under-
standing and applying the following key success factors and strategies
can accelerate your company’s widescale adoption of virtualization.
YO U R V I S I O N F O R S U CC E S S
Top-down sponsorship ensures
success of virtualization implementation
Executive sponsorship and technical champions ensure
the appropriate levels of funding, staffing, and cross-group
cooperation and communication.
•
CIOCIO
IT
IT
IT
CIO
Line ofIT
Line ofBusiness
Line ofBusiness
Line ofBusiness
Treat virtualization as an architectural
decision throughout the organization
By regarding virtualization as an architectural decision,
you can enable, accelerate and save costs for multiple
projects and business initiatives. This standardization
can help achieve the maximum ROI for your organization.
•
V I S I O NS TR AT E G Y
K E Y CO N S I D E R AT I O N SPA S S D O W N
Achieve stakeholder buy-in early & maintain through all phases
One of your challenges may be the cultural resistance encountered when
new technology is introduced. This can be easily overcome by selling
stakeholders on added benefits, higher levels of service, and a new
service delivery model that improves the time to provision computing
power from months to minutes.
•
Design for the big picture, but deploy
incrementally – target initial ROI within six months
Organizations can reap rapid return on investment benefits by
deploying projects incrementally and applying the learning from
each incremental project to the next. Virtualization maturity and
confidence develop rapidly this way. Many VMware enterprise
customers have started with a server consolidation project
and rapidly used this success to evolve into a standardized
implementation with more mature management processes.
•
Start your initial project with an
infrastructure assessment and design review
Upfront assessment will ensure high quality design and enables
you to migrate to a virtualized environment in a controlled
and predictable way with proper remediation and change
management procedures. This groundwork will ensure an
implementation plan that is specific to your organization’s
infrastructure environment and needs.
•
Form a virtualization core team to be the
initial agents of change, ensuring early success
Initiating a virtual infrastructure core team or “SWAT” team formed
from the IT infrastructure ranks has been the hallmark of all
successful, broad scale deployments that VMware has observed.
•
proof ofconcept
departmentrollout
expandedrollout
process andtechnologystandardized
standardize expandadopt
numberof VM’s
too muchtoo soon
too littletoo late
time & virtual maturity
�
N OT E S
Transition core team to an operational center of excellence
Once your first project ramps up and the benefits of
virtualization are proven, multiple parts of your enterprise
will start to request virtualization for their projects. To
accommodate this demand, your core team should refine
and standardize the processes established in the initial
project.
•
proof ofconcept
departmentrollout
expandedrollout
process andtechnologystandardized
standardize expandadopt
numberof VM’s
too muchtoo soon
too littletoo late
time & virtual maturity
V I S I O NS TR AT E G Y
K E Y CO N S I D E R AT I O N SPA S S D O W N
QUESTION
WhAT ARE ThE TOP 3 BUSINESS DRIVERS FOR VIRTUALIzATION IN YOUR COmPANY?
1.
2.
3.
Major data center infrastructure consoli-dation, migration or change events
Branch office consolidation
Cost-avoidance of data center facilities (space, power, cooling, etc)
Improved business continuity and appli-cation protection, in-sourcing of disaster recovery
Meeting end-user environment security, compliance goals, particularly as part of major desktop refresh
Accelerating end-user enablement for outsourcing and off-shoring initiatives
Accelerating large-scale application development, upgrades and implemen-tation projects
Accelerated development and test envi-ronment due to ease of implementation and minimal risk to the enterprise envi-ronment
Server consolidation targets the migration and re-hosting of existing, underutilized infrastructure in the data center, driven by the need to reduce cost, space and other
•
•
•
•
•
•
•
•
•
environment and capacity constraints
Server containment is a more rapid imple-mentation strategy used to virtualize most or all new applications in a properly sized virtual infrastructure, independent of whether existing underutilized assets have been virtualized or consolidated.
Business continuity/disaster recovery expansion. The cost-reduced and sim-plified disaster recovery methods avail-able in a virtual environment can enable expansion of disaster recovery to new applications, or in sourcing of a DR plan. As a data point, 63% of VMware custom-ers have used VMware to enhance disaster recovery capabilities.
Virtual desktop infrastructure allows centralized consolidation, backup and business continuity options for end-user environments hosted in a secure data center, maps particularly well to the need to centrally secure, manage and rapidly provision remote or 3rd party knowledge workers.
•
•
•
KEy bUSINESS DRIVERS
9
N OT E S
V I S I O NS TR AT E G y
K E Y CO N S I D E R AT I O N SPA S S D O W N
QUESTIONS
STRATEGIC QUESTIONS TO BE ANSWERED
How do I effectively sponsor the enterprise vir tualization vision within the company and keep the organization focused through all stages of adoption?
How do I obtain and maintain line of business stakeholder buy-in and support for vir tualization throughout all stages of implementation?
How do I slip-stream vir tualization into existing projects to accelerate adoption without introducing risk?
What is the changing role of the OS within my enterprise over the next 18-24 months?
•
•
•
•
W h AT A R E T h E m O S T U R G E N T Q U E S T I O N S A N E X E C U T I V E S h O U L D A N S W E R W h E N D E V E LO P I N G A P L A N TO I m P L E m E N T V I R T UA L I z AT I O N AC R O S S T h E CO m PA N Y ?
V I S I O NS TR AT E G y
K E Y CO N S I D E R AT I O N SPA S S D O W N
2 - 6 weeks 2 - 4 weeks 4 - 8 weeks On-going
Assess Planning andDesign Build Manage
• Determine operational readiness
• Document current environment
• Review current/planned projects
• Document current application portfolio
• Develop operational designs
• Conduct training
• De�ne service o�erings
• Develop a test plan
• Execute the blueprints and test plans
• Integrate into the organization
• Update process guides and procedures
• P2V Migrations
• Go live
• Continuous Improvement
After completing an initial project to validate functionality,
there are some practical next steps to ensure successful deploy-
ment in all future projects. This Executive workbook focuses primarily
on the “Build VCoE”, “Assess”, “Plan and Design” phases. Future tools
wil l provide additional details of the “Build” and “Manage” phases.
The duration of each phase can vary based on the scope of the
project, expertise, resource availabil ity, and organizational readiness.
Vir tualization capabil it ies should be developed in parallel through-
out all phases.
G E T T I N G S TA R T E D : B E YO N D T h E I N I T I A L P R O j E C T
11
Establish your vision for vir tualization and share ROI from initial project
Build a Virtualization Center of Excellence (VCoE)
Determine level of operational readiness as compared to your enterprise vir tualization vision
Assess and document IT infrastructure environment
Assess current and planned projects for impact and fit for vir tualization
Inventory server assets and assess server util ization
Document your application portfolio’s fit with vir tualization
Plan and design
Build and deploy incrementally
Manage and optimize
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
TEN EASy STEPS TO VIRTUAlIzATION
N OT E S
V I S I O NS TR AT E G y
K E Y CO N S I D E R AT I O N SPA S S D O W N
STORAGE
FINANCE
NETWORK
OPERATIONS
ChANGE
MANAGEMENT
CAPACITy
MANAGEMENT
SERVER
SUPPORT
VCoEVirtual Infrastructure Center
of Excellence
V I R T UA L I z AT I O N C E N T E R O F E XC E L L E N C E
After the initial project has been deployed, the core team or SWAT team
used to deploy the initial project wil l have some key learning to impart
to the organization. Therefore, in parallel with the next project, it is
t ime to transition the core team function to a Vir tualization Center of
Excellence ( VCoE). One recommendation for how to execute this is that
each key functional group nominates a respected representative to the
VCoE, establishing a feedback loop between the VCoE and each group.
Information is shared bi-directionally, ensuring cross-functional support
and growth. Representatives will drive new ideas, resolve operational
problems, and create innovative vir tualization solutions to ensure
continuous process improvement.
S A M P l E V C OE
V I S I O NS TR AT E G y
K E Y CO N S I D E R AT I O N SPA S S D O W N
technologypeople process
steps to maturity
initial repeatable de�ned managed optimized
TIPPING POINT
I D E N T I F Y I N G C U R R E N T A N D D E S I R E DV m WA R E m AT U R I T Y m O D E L
In the “physical” world, there are built-in latencies in provisioning. The goal of the VCoE is to accelerate and ensure inter-group communica-tion and to take advantage of the increased speed of change that vir tual-ization allows. The VCoE provides a f lexible framework for organizational re-alignment – proactively responding to changing business needs.
The VCoE ensures vir tual infrastructure support teams are adequately staffed and trained to take full advantage of the positive changes that vir tualization brings to the organization.”
15
N OT E S
technologypeople process
steps to maturity
initial repeatable de�ned managed optimized
TIPPING POINT
Initial - The IT service delivery process is characterized as ad hoc, and occasionally even chaotic. Few processes are defined, and success depends on individual effort and heroics.
Repeatable - Basic service management processes are established. The necessary discipline is in place to repeat earlier successes on similar services with similar service levels.
Defined - The IT service processes are documented, standardized, and integrated into standard service processes. All services are delivered using approved, tailored versions of the organization’s standard service processes.
Managed - Detailed measurements of the IT service delivery process and service quality are collected. Both the service processes and the delivery services are quantitatively understood and controlled.
Optimizing - Continuous process improvement is enabled by quantitative feedback from the processes and from piloting innovative ideas and technologies.
1.
2.
3.
4.
5.
VMWARE MATURITy MODEl (STAGES 1-5)
V I S I O NS TR AT E G y
K E Y CO N S I D E R AT I O N SPA S S D O W N
CIOCIO
IT
IT
IT
CIO
Line ofIT
Line ofBusiness
Line ofBusiness
Line ofBusiness
C R I T I C A L S U CC E S S FAC TO R S S U m m A R Y
Top-down sponsorship ensures success of vir tualization
implementations
Treat vir tualization as an architectural decision, design for the big
picture, but deploy incrementally targeting ROI within six months
Form a vir tualization core team to be the agents of change, ensuring
early success
Ensure high quality design and remediation to avoid early shutdown
Achieve stakeholder buy-in early and maintain through all phases
Socialize benefits of vir tualization to business owners by selling:
Faster provisioning of applications
Higher availability and dynamic resource management
Accelerate application projects with increased development
lifecycle productivity
•
•
•
•
•
•
»
»
»
19
B U I L D V I R T UA L I z AT I O N C E N T E R O F E XC E L L E N C E ( V CO E )
VMware has a created a model for a typical Virtualization CoE team that describes the roles, skill sets, and background desired for such a team. Most organizations customize or map these prototype recommendations into skills and roles that are appropriate to their own organization.
Once internal education on the benefits of optimizing system infrastructure
through vir tualization ramps up, multiple parts of your enterprise wil l star t to
request vir tualization for their projects. To prepare for this demand, refine the
processes established in the initial project and transition your core team sk il ls
to more standardized and measurable processes. Many companies call this
phase of vir tual infrastructure learning a “center of excellence” (CoE) which
has its own business and service delivery inter face, technical design, and
operational competence.
CIO
Virtualization Center of
Excellence
CIO
CentralIT
Organizational Alignment
DataCenter 2
Line ofBusiness IT
Line ofBusiness IT
Line ofBusiness IT
Line ofBusiness IT
Line ofBusiness IT
Line ofBusiness IT
Production Systemsand Infrastructure
HA, DR Support
ERP
CRM
SFA
HRMS
B2B
PROCUREMENT
Vir
tual
izat
ion
Infr
astr
uct
ure
Virtualization Center of
Excellence
Finance
Lines ofBusiness
CustomerService
Customer Service
HumanResources
Sales
WindowsOS Admin
Linux Ops
UNIXEngineering
Procurement
Security
Network
Storage
V I S I O NS TR AT E G Y
K E y CO N S I D E R AT I O N SPA S S D O W N
Fill in names that best f it members of your organization:
Engineering
Support
Procurement
Storage
Finance
Capacity Management
Data Management
Server Maintenance
Security
Patch Management
Change Management
Network Operations
Network Design
Data Center
Operations Staff
Total Head Count
VCoE TEAm STRUCTURE
Team lead
Architect
Analyst
Administrator
Administrator
V I S I O NS TR AT E G Y
K E y CO N S I D E R AT I O N SPA S S D O W N
A vir tualized infrastructure can introduce changes (many of them
highly beneficial) to the IT infrastructure that can be readily
addressed through proper education and co-operative planning
within the various IT infrastructure practices. This table provides a
bird’s eye view of the nature of the changes, and some consider-
ations for how and where to address the impact of changes.
P h A S E 1: A S S E S S , E D U C AT E , P L A N
UNDERSTANDINg CHANgES TO THE REST OF THE IT INFRASTRUC TURE
Area Key Changes and Considerations Recommendations Notes for Executive:
Dat
acen
ter
Faci
liti
es
Dramatic reduction and avoidance in aggregate power, cooling consumption, and space requirements
70-80% reduction in many cases
Slight power and thermal density increases possible due to more highly utilized, dense memory systems
Density increases should be well within headroom associated with current datacenter designs, (unlike prior blade systems)
•
-
•
-
Build hard operational savings into ROI, if needed. Most VMware and our partner ROI tools take this into account
Utilize the output of the VI hardware design and system vendor calculators and specifications to ensure compliance with your datacenter specs
•
•
Ind
ust
ry S
tan
dar
d S
erve
r h
ard
war
e D
esig
n
Full, larger memory, higher availability I/O configurations over prior x86 designs
1-2 standardized design configurations for VI vs. application-specific dedicated hardware
•
•
Operations/facilities teams should be educated and to verify the feasibility of these designs
Ensure procurement process/teams understand “standardized” VI design vs. application specific physical HW requests
•
•
Shar
ed
Sto
rag
e
Enterprise deployments will generally dictate a shared storage architecture
FC or IP SAN, NAS based on cost and environment
“Port consolidation” makes SAN attach of new workloads on VI a lot more cost-effective
Dedicated, standalone SAN or leverage of corporate SAN for VI
Requires SAN team to treat VI “cluster” as new type of host
•
-
•
•
•
Implement refinements to existing standards & procedures specific to VI with storage eng/ops teams:
Mapping of VI storage requirements to SAN storage tiers
Batch pre-provisioning of lUNs to VI team vs. app- or server-specific requests
New SAN zoning & lUN configuration guidelines
•
-
-
-
21
Area Key Changes and Considerations Recommendations Notes for Executive:
Dat
acen
ter
Faci
liti
es
Dramatic reduction and avoidance in aggregate power, cooling consumption, and space requirements
70-80% reduction in many cases
Slight power and thermal density increases possible due to more highly utilized, dense memory systems
Density increases should be well within headroom associated with current datacenter designs, (unlike prior blade systems)
•
-
•
-
Build hard operational savings into ROI, if needed. Most VMware and our partner ROI tools take this into account
Utilize the output of the VI hardware design and system vendor calculators and specifications to ensure compliance with your datacenter specs
•
•
Ind
ust
ry S
tan
dar
d S
erve
r h
ard
war
e D
esig
n
Full, larger memory, higher availability I/O configurations over prior x86 designs
1-2 standardized design configurations for VI vs. application-specific dedicated hardware
•
•
Operations/facilities teams should be educated and to verify the feasibility of these designs
Ensure procurement process/teams understand “standardized” VI design vs. application specific physical HW requests
•
•
Shar
ed
Sto
rag
e
Enterprise deployments will generally dictate a shared storage architecture
FC or IP SAN, NAS based on cost and environment
“Port consolidation” makes SAN attach of new workloads on VI a lot more cost-effective
Dedicated, standalone SAN or leverage of corporate SAN for VI
Requires SAN team to treat VI “cluster” as new type of host
•
-
•
•
•
Implement refinements to existing standards & procedures specific to VI with storage eng/ops teams:
Mapping of VI storage requirements to SAN storage tiers
Batch pre-provisioning of lUNs to VI team vs. app- or server-specific requests
New SAN zoning & lUN configuration guidelines
•
-
-
-
V I S I O NS TR AT E G Y
K E y CO N S I D E R AT I O N SPA S S D O W N
P h A S E 2: P L A N A N D D E S I G N
Area Key Changes and Considerations Recommendations Notes for Executive:
Net
wo
rk I
nfr
astr
uct
ure
Simplification and cost savings from network port and cable consolidation
larger # of higher speed ports to each server (vs. prior x86 designs)
Specialized needs for VMotion and management network segments
•
•
•
Educate and establish appropriate refinements to standards and operating procedures with network engineering/operations teams:
New standard designs for VI configurations
Batch provisioning of IPs to VI team
Address need for DHCP assignment to servers for certain use models
•
-
-
-
Syst
em a
nd
Ap
pli
cati
on
Ava
ilab
ilit
y
Overall increased availability from shared storage and higher availability HW design
Operational flexibility and greatly minimized downtime for planned maintenance ( VMotion)
larger risk domain in case of unplanned single system hardware failure
•
•
•
Implement built-in ( VMware HA) or 3rd party VI availability capabilities
Re-visit which applications are clustered during application candidate selection
Document “virtual system” availability service agreements as appropriate
•
•
•
bac
kup
an
d S
yste
m R
eco
very
Standard backup and recovery tools and processes can be translated to virtual infrastructure with minimal disruption
Scheduling rules and capacity headroom for standard network backup agents need to be accommodated
VI enables more efficient strategies for applying a new set of “off-VM” techniques that can improve backup/recovery service levels different workloads
•
-
•
Pull-in backup and recovery ops teams into design and deployment planning sessions
Follow VMware best practices for tiering and integrating new backup options
Consider org readiness to consider new data protection tooling for VI, and fold in new backup options (e.g. VMware VCB) using existing and/or VMware-specific data protection tools from 3rd parties
•
•
•
2�
P h A S E 2: P L A N A N D D E S I G N
Area Key Changes and Considerations Recommendations Notes for Executive:
Net
wo
rk I
nfr
astr
uct
ure
Simplification and cost savings from network port and cable consolidation
larger # of higher speed ports to each server (vs. prior x86 designs)
Specialized needs for VMotion and management network segments
•
•
•
Educate and establish appropriate refinements to standards and operating procedures with network engineering/operations teams:
New standard designs for VI configurations
Batch provisioning of IPs to VI team
Address need for DHCP assignment to servers for certain use models
•
-
-
-
Syst
em a
nd
Ap
pli
cati
on
Ava
ilab
ilit
y
Overall increased availability from shared storage and higher availability HW design
Operational flexibility and greatly minimized downtime for planned maintenance ( VMotion)
larger risk domain in case of unplanned single system hardware failure
•
•
•
Implement built-in ( VMware HA) or 3rd party VI availability capabilities
Re-visit which applications are clustered during application candidate selection
Document “virtual system” availability service agreements as appropriate
•
•
•
bac
kup
an
d S
yste
m R
eco
very
Standard backup and recovery tools and processes can be translated to virtual infrastructure with minimal disruption
Scheduling rules and capacity headroom for standard network backup agents need to be accommodated
VI enables more efficient strategies for applying a new set of “off-VM” techniques that can improve backup/recovery service levels different workloads
•
-
•
Pull-in backup and recovery ops teams into design and deployment planning sessions
Follow VMware best practices for tiering and integrating new backup options
Consider org readiness to consider new data protection tooling for VI, and fold in new backup options (e.g. VMware VCB) using existing and/or VMware-specific data protection tools from 3rd parties
•
•
•
V I S I O NS TR AT E G Y
K E y CO N S I D E R AT I O N SPA S S D O W N
P h A S E 2: P L A N A N D D E S I G N
Area Key Changes and Considerations Recommendations Notes for Executive:
Dis
aste
r R
eco
very
VI enables new cost structure and automation capabilities to dramatically simplify and expand DR coverage
Significantly better cost, SlO, RPO, RTO and capex compared to 1:1 physical designs, or outsourced service options
Up front design decisions (e.g. SAN capabilities, mixing production and dev/test workloads) can optimize the cost and efficiency of a subsequent DR/BC design
Certain organizations may be willing to use a “physical to virtual” DR strategy to implement expanded DR to workloads and applications that will not be initially virtualized
•
-
•
•
Use improved or expanded BC/DR beyond the hard cost savings ROI to augment the rationale and organization buy-in for rolling out VI
Roll in high level BC/DR considerations (but not a complete design) into Phase 1
Typically, most organizations roll out a compete design and implementation for a “V-to-V” BC/DR strategy as part of a Phase 2 deployment, as baseline operational skills around a single site VI is a pre-requisite
Evaluate “P2V DR” strategy if organizational resistance or priorities dictate such a path
•
•
-
•
Co
st A
llo
cati
on
/Ch
arg
ebac
k
VI reduces infrastructure costing for server workloads compared to a physical server environment. Commonly in the range of :
60% less for initial deployment
10:1 or greater cost savings for hard on-going infrastructure costs
Offers a cost-justified way to introduce dramatically lower cost service tiers for infrastructure to business units
•
-
-
-
If a chargeback model exists, create a new service tier that roughly represents the new cost structure (e.g. 20% of existing physical server cost). This will provide additional rationale for business owner buy-in to VI
If a chargeback model does not exist, use VI to introduce such a model as part of a Phase 2 deployment
E.g. 2-3 tiered service offerings for virtual/physical server & storage infrastructure
Focus on simple, cost-based allocations independent of usage. Defer sophisticated “metering” or usage based models well into or beyond Phase 2 deployments
•
•
-
•
25
P h A S E 2: P L A N A N D D E S I G N
Area Key Changes and Considerations Recommendations Notes for Executive:
Dis
aste
r R
eco
very
VI enables new cost structure and automation capabilities to dramatically simplify and expand DR coverage
Significantly better cost, SlO, RPO, RTO and capex compared to 1:1 physical designs, or outsourced service options
Up front design decisions (e.g. SAN capabilities, mixing production and dev/test workloads) can optimize the cost and efficiency of a subsequent DR/BC design
Certain organizations may be willing to use a “physical to virtual” DR strategy to implement expanded DR to workloads and applications that will not be initially virtualized
•
-
•
•
Use improved or expanded BC/DR beyond the hard cost savings ROI to augment the rationale and organization buy-in for rolling out VI
Roll in high level BC/DR considerations (but not a complete design) into Phase 1
Typically, most organizations roll out a compete design and implementation for a “V-to-V” BC/DR strategy as part of a Phase 2 deployment, as baseline operational skills around a single site VI is a pre-requisite
Evaluate “P2V DR” strategy if organizational resistance or priorities dictate such a path
•
•
-
•
Co
st A
llo
cati
on
/Ch
arg
ebac
k
VI reduces infrastructure costing for server workloads compared to a physical server environment. Commonly in the range of :
60% less for initial deployment
10:1 or greater cost savings for hard on-going infrastructure costs
Offers a cost-justified way to introduce dramatically lower cost service tiers for infrastructure to business units
•
-
-
-
If a chargeback model exists, create a new service tier that roughly represents the new cost structure (e.g. 20% of existing physical server cost). This will provide additional rationale for business owner buy-in to VI
If a chargeback model does not exist, use VI to introduce such a model as part of a Phase 2 deployment
E.g. 2-3 tiered service offerings for virtual/physical server & storage infrastructure
Focus on simple, cost-based allocations independent of usage. Defer sophisticated “metering” or usage based models well into or beyond Phase 2 deployments
•
•
-
•
V I S I O NS TR AT E G Y
K E y CO N S I D E R AT I O N SPA S S D O W N
S U m m A R Y A N D K E Y TA K E AWAYS
Have your organizations understand that vir tualization is an architectural decision and a new IT service—not another new project.
Appoint a high-caliber internal team and empower them to learn, design, operate, and drive internal changes to make vir tual infrastructure successful.
Focus heavily on business owners buy-in, particularly during the initial stages. Educate business owners on improved benefits and service levels, engage them actively in acceptance testing, and provide quality remediation options.
Secure a proper upfront assessment of existing application and assets, and a high quality, long term design of the vir tual infrastructure. Involve and educate internal IT stakeholders throughout the design and rollout process.
Use proper project phasing and workload selection to quickly achieve initial success. Reinvest the organizational confidence derived from the first pilot to expand deployments.
Evolve your core team into a VCoE to provide governance as well as to maintain and manage your vir tual infrastructure over time.
•
•
•
•
•
•
2�
TImE TO VALUE AND ROI
N OT E S
scope
POC Phase 1 Phase 2
usage
T I M E - T O - V A L U E
• limited scope• minimal process change• organizational learning• tier 2 workloads
• test & dev • containment
• disaster recovery• consolidation migration
• full standardization• process optimization
• expansion• key IT process integration• multiple LOB• tier 1 workloads
29
V I S I O NS TR AT E G Y
K E Y CO N S I D E R AT I O N SPA S S D O W N
Phased deployment are focused on server containment and
consolidation. First, use an upfront design and assessment as
well as a quick Phase 1 deployment to rapidly scale competencies.
Defer more sophisticated and optimized capabilities to
subsequent deployments.
While the goal of vir tualization is to improve operations and reduce
cost, the best practice in deploying vir tualization is to take a phased
implementation approach where you focus on the following areas:
Early ROI: Choose first phase candidate workloads that are
low risk, high visibility and generate early ROI.
low Risk: Scope initial projects appropriately to mitigate
any inherent risks in deployment.
Critical learning: Build confidence in organizational
capabilities by communicating learning and success.
By using a phased approach initial benefits can be realized much
more quick ly by work ing initial ly with a business partner most suited
to recognize the value of vir tualization. Use the first phase to demon-
strate the validity and benefits of vir tualization to more skeptical
business partners.
•
•
•
E X E C U T I V E PA S S D O W N
N OT E S
V I S I O NS TR AT E G Y
K E Y CO N S I D E R AT I O N SPA S S D O W N
QUESTIONS
TAC T I C A L Q U E S T I O N S TO B E A N S W E R E DHow do I select and empower a high-caliber SWAT Team to get started?
When is the right time to establish a Vir tualization Center of Excellence?
How do I measure and maintain vir tualization expertise over time?
How do I measure my organization’s vir tual maturity?
How do I obtain ISV vendors support for VM-based license pricing?
How do I implement chargeback mechanisms?
What are the best workloads to vir tualize first?
What are right production systems to vir tualize?
Do I have a disaster recovery and “anytime” backup strategy that leverages vir tualization?
•
•
•
•
•
•
•
•
•
V I S I O NS TR AT E G Y
K E Y CO N S I D E R AT I O N SPA S S D O W N
P h A S E 0: B U I L D V I R T UA L I z AT I O N C E N T E R O F E XC E L L E N C EThis table provides additional information to enable IT Staff to get started by collecting data as part of the planning and assessment steps.
Role Key Responsibilities Datacenter VP/Director
Rel
atio
nsh
ip M
anag
er
Understand and keep up-to-date with current and expected future business needs.
Provide guidance to application developers on the impact of virtualization to their work.
IT A
nal
yst
Understand how each Service Management function is impacted by virtualization.
Recommend changes to operational processes to support new virtualized projects and applications (for example, new application deployment model, ongoing capacity management, etc.).
Develop the organization's strategy and roadmap for deploying virtualization.
IT I
nfr
astr
uct
ure
A
rch
itec
t
Translate business requirements into architectural designs.
Direct development and maintenance of virtual infrastructure blueprints and documentation.
Develop virtual infrastructure patching policies and procedures.
Establish credibility by developing VMware architecture expertise.
IT I
nfr
astr
uct
ure
En
gin
eer
Provide specific technical designs for virtualized solutions.
Install virtualization software.
Execute test plans.
Develop virtual environment management guide.
��
Role Key Responsibilities Datacenter VP/Director
Rel
atio
nsh
ip M
anag
er
Understand and keep up-to-date with current and expected future business needs.
Provide guidance to application developers on the impact of virtualization to their work.
IT A
nal
yst
Understand how each Service Management function is impacted by virtualization.
Recommend changes to operational processes to support new virtualized projects and applications (for example, new application deployment model, ongoing capacity management, etc.).
Develop the organization's strategy and roadmap for deploying virtualization.
IT I
nfr
astr
uct
ure
A
rch
itec
t
Translate business requirements into architectural designs.
Direct development and maintenance of virtual infrastructure blueprints and documentation.
Develop virtual infrastructure patching policies and procedures.
Establish credibility by developing VMware architecture expertise.
IT I
nfr
astr
uct
ure
En
gin
eer
Provide specific technical designs for virtualized solutions.
Install virtualization software.
Execute test plans.
Develop virtual environment management guide.
V I S I O NS TR AT E G Y
K E Y CO N S I D E R AT I O N SPA S S D O W N
Key Responsibilities Datacenter VP/Director
Project name and description
Project manager
business sponsor
Estimated project completion date and key milestones
Project status
Project size (or estimated size)
Anticipated IT infrastructure needs
how project is impacted by virtualization (initial determination to be done by project manager or project team and validated by CoE team)
P h A S E 1: I D E N T I F Y C U R R E N T A N D P L A N N E D P R O j E C T S W I T h F I T F O R V I R T UA L I z AT I O N ( T h E S E A R E S L I P - S T R E A m C A N D I D AT E S )
Identify candidate projects where there may be particular synergies with
vir tualization, for example, a major application release, IT infrastructure
refresh, or physical data center consolidation. To prioritize the effor t,
a minimum threshold project size (for example, >$100K) can be set to
identify the most important projects.
�5
Key Responsibilities Datacenter VP/Director
Project name and description
Project manager
business sponsor
Estimated project completion date and key milestones
Project status
Project size (or estimated size)
Anticipated IT infrastructure needs
how project is impacted by virtualization (initial determination to be done by project manager or project team and validated by CoE team)
V I S I O NS TR AT E G Y
K E Y CO N S I D E R AT I O N SPA S S D O W N
Environment Data Datacenter VP/Director
Type of IT infrastructure asset (for example, server, storage)
Vendor
hardware configuration (for example, DP Dl�80 2-way, EMC Symmetrix)
Operating System
Asset physical location
Applications that currently use the IT infrastructure asset
Current asset utilization (for example, peak and average utilization over a �0-day period)
Associated infrastructure (for example, physical network, shared versus local storage)
P h A S E 1: D O C U m E N T C U R R E N T E N V I R O N m E N T, I N V E N TO R Y A S S E T S & S E R V E R U T I L I z AT I O N
��
Environment Data Datacenter VP/Director
Type of IT infrastructure asset (for example, server, storage)
Vendor
hardware configuration (for example, DP Dl�80 2-way, EMC Symmetrix)
Operating System
Asset physical location
Applications that currently use the IT infrastructure asset
Current asset utilization (for example, peak and average utilization over a �0-day period)
Associated infrastructure (for example, physical network, shared versus local storage)
V I S I O NS TR AT E G Y
K E Y CO N S I D E R AT I O N SPA S S D O W N
Application Info Datacenter VP/Director
Application name
Key user groups
Dependencies on other applications
Where application is in it’s lifecycle (for example, scheduled to be retired in 6 months, scheduled for major upgrade in one year)
licensing agreements (for packaged software)
Supporting infrastructure (for example, server type or storage type)
level of I/O (can rate as “low”, “medium”, or “high”)
Suitability for migration to virtual infrastructure (based on above criteria)
P h A S E 1: A S S E S S A P P L I C AT I O N P O R T F O L I O F I T W I T h V I R T UA L I z AT I O N
�9
Application Info Datacenter VP/Director
Application name
Key user groups
Dependencies on other applications
Where application is in it’s lifecycle (for example, scheduled to be retired in 6 months, scheduled for major upgrade in one year)
licensing agreements (for packaged software)
Supporting infrastructure (for example, server type or storage type)
level of I/O (can rate as “low”, “medium”, or “high”)
Suitability for migration to virtual infrastructure (based on above criteria)
P h A S E 1: A S S E S S A P P L I C AT I O N P O R T F O L I O F I T W I T h V I R T UA L I z AT I O N
V I S I O NS TR AT E G Y
K E Y CO N S I D E R AT I O N SPA S S D O W N
Technology Area Design and Operational Considerations Datacenter VP/Director Virtualization Center of Excellence
Distributed topology and high-level design
Configuration of physical clusters/farms and logical resource pools
Deployment topology, security of VC management servers
Standardized HW recommendations/designs for server infrastructure
•
•
•
SAN and Networking Integration Key design standards and constraints for key interfaces with other corporate resource teams
•
Workload and Guest Design and Management
workload selection process
application and guest access
•
•
Standard OS-specific guest configurations, agents and templates
Target VM density for different workloads/ workload classes
Virtual Workload Selection/Admission Process
Administration/Access Roles and Rights to rest of IT and lOB users
guest-specific backup, recovery and protection
•
•
•
•
•
VMware Infrastructure Configuration and Operations
backup and recovery
resource management
standardized configuration
•
•
•
Standardized builds and configurations for VMware ESX Server
Deployment, maintenance and recovery of the VMware infrastructure itself
•
•
P h A S E 2: T E C h N I C A L A N D O P E R AT I O N A L P L A N N I N G A N D D E S I G N
Roll ing out vir tual infrastructure provides an opportunity for you to develop
new technical sk il ls and design expertise specific to this infrastructure.
The table below highlights areas where core teams focused on VMware infra-
structure will need to develop deployment, training and operational plans.
41
Technology Area Design and Operational Considerations Datacenter VP/Director Virtualization Center of Excellence
Distributed topology and high-level design
Configuration of physical clusters/farms and logical resource pools
Deployment topology, security of VC management servers
Standardized HW recommendations/designs for server infrastructure
•
•
•
SAN and Networking Integration Key design standards and constraints for key interfaces with other corporate resource teams
•
Workload and Guest Design and Management
workload selection process
application and guest access
•
•
Standard OS-specific guest configurations, agents and templates
Target VM density for different workloads/ workload classes
Virtual Workload Selection/Admission Process
Administration/Access Roles and Rights to rest of IT and lOB users
guest-specific backup, recovery and protection
•
•
•
•
•
VMware Infrastructure Configuration and Operations
backup and recovery
resource management
standardized configuration
•
•
•
Standardized builds and configurations for VMware ESX Server
Deployment, maintenance and recovery of the VMware infrastructure itself
•
•
V I S I O NS TR AT E G Y
K E Y CO N S I D E R AT I O N SPA S S D O W N
Area Key Changes and Consideration Recommendations Datacenter VP/Director Virtualization Center of Excellence
Cos
t A
lloca
tion
/Ch
arg
ebac
k
VI reduces infrastructure costing for server workloads compared to a physical server environment. Commonly in the range of :
60% less for initial deployment
10:1 or greater cost savings for hard on-going infrastructure costs
Offers a cost-justified way to introduce dramatically lower cost service tiers for infrastructure to business units
•
-
-
-
If a chargeback model exists, create a new service tier that roughly represents the new cost structure (e.g. 20% of existing physical server cost). This will provide additional rationale for business owner buy-in to VI.
If a chargeback model does not exist, use VI to introduce such a model as part of a Phase 2 deployment
E.g. 2-3 tiered service offerings for virtual/physical server & storage infrastructure
Focus on simple, cost-based allocations independent of usage. Defer sophisticated “metering” or usage based models well into or beyond Phase 2 deployments.
•
•
-
•
Win
dow
s, l
inux
an
d A
pp
licat
ion
Tea
m In
teg
rati
on
Many have found virtual infrastructure implementations to be a catalyst for increased teamwork, integration or outright unification of “OS and application stack” silos in the IT organization.
Windows and linux teams should view their interface with the virtual infrastructure core team as the providers of their “hardware infrastructure”.
Outside of initial application and OS sizing/base-lining/provisioning, system administrator’s day-to-day tasks and expertise should be minimally affected.
Areas of impact and training typically include:
Virtual infrastructure team should be consulted for incident resolution around application performance & service levels.
Standard OS-centric monitoring and performance tools will need “re-interpretation” in a VI implementation.
•
•
•
•
•
•
Pull in Windows, Solaris and linux and other OS engineering and ops teams as appropriate into design and deployment planning sessions.
Modify only the minimum set of roles and tools to preserve the tools and skill sets of existing OS admin teams to interface with the VI team as a hardware/infrastructure team. i.e. OS admin roles and tooling should be preserved for:
Patch and OS management
OS & app-specific problem/incident management
Application deployment/changes
Access methods and security rules for virtual machine access by OS, system administrators and application teams need to be defined and implemented.
•
•
-
-
-
•
Prob
lem
, Cha
nge
and
Inci
dent
Mgt
The virtual infrastructure layer and CoE team should be carefully considered to see if they need to be in the monitoring/change and remediation path for established IT operations teams and processes around:
Fault/availability events in HW or at the application level
Configuration changes to OS, network, storage infrastructure.
Application performance
Infrastructure (e.g. HW ) service and changes
•
-
-
-
-
Preserve as many of the problem/incident monitoring and resolution workflows, org responsibilities as possible.
Ensure a responsive, technically educated SWAT team for virtual infrastructure problem remediation, particularly during the first phase of deployment to achieve business and client buy-in.
Implement off-the-shelf problem and reporting modules that fit into your current monitoring, problem resolution tooling and are “VMware-aware”.
•
•
•
P h A S E 2 - 4: P L A N , D E S I G N , B U I L D A N D m A N AG E
4�
Area Key Changes and Consideration Recommendations Datacenter VP/Director Virtualization Center of Excellence
Cos
t A
lloca
tion
/Ch
arg
ebac
k
VI reduces infrastructure costing for server workloads compared to a physical server environment. Commonly in the range of :
60% less for initial deployment
10:1 or greater cost savings for hard on-going infrastructure costs
Offers a cost-justified way to introduce dramatically lower cost service tiers for infrastructure to business units
•
-
-
-
If a chargeback model exists, create a new service tier that roughly represents the new cost structure (e.g. 20% of existing physical server cost). This will provide additional rationale for business owner buy-in to VI.
If a chargeback model does not exist, use VI to introduce such a model as part of a Phase 2 deployment
E.g. 2-3 tiered service offerings for virtual/physical server & storage infrastructure
Focus on simple, cost-based allocations independent of usage. Defer sophisticated “metering” or usage based models well into or beyond Phase 2 deployments.
•
•
-
•
Win
dow
s, l
inux
an
d A
pp
licat
ion
Tea
m In
teg
rati
on
Many have found virtual infrastructure implementations to be a catalyst for increased teamwork, integration or outright unification of “OS and application stack” silos in the IT organization.
Windows and linux teams should view their interface with the virtual infrastructure core team as the providers of their “hardware infrastructure”.
Outside of initial application and OS sizing/base-lining/provisioning, system administrator’s day-to-day tasks and expertise should be minimally affected.
Areas of impact and training typically include:
Virtual infrastructure team should be consulted for incident resolution around application performance & service levels.
Standard OS-centric monitoring and performance tools will need “re-interpretation” in a VI implementation.
•
•
•
•
•
•
Pull in Windows, Solaris and linux and other OS engineering and ops teams as appropriate into design and deployment planning sessions.
Modify only the minimum set of roles and tools to preserve the tools and skill sets of existing OS admin teams to interface with the VI team as a hardware/infrastructure team. i.e. OS admin roles and tooling should be preserved for:
Patch and OS management
OS & app-specific problem/incident management
Application deployment/changes
Access methods and security rules for virtual machine access by OS, system administrators and application teams need to be defined and implemented.
•
•
-
-
-
•
Prob
lem
, Cha
nge
and
Inci
dent
Mgt
The virtual infrastructure layer and CoE team should be carefully considered to see if they need to be in the monitoring/change and remediation path for established IT operations teams and processes around:
Fault/availability events in HW or at the application level
Configuration changes to OS, network, storage infrastructure.
Application performance
Infrastructure (e.g. HW ) service and changes
•
-
-
-
-
Preserve as many of the problem/incident monitoring and resolution workflows, org responsibilities as possible.
Ensure a responsive, technically educated SWAT team for virtual infrastructure problem remediation, particularly during the first phase of deployment to achieve business and client buy-in.
Implement off-the-shelf problem and reporting modules that fit into your current monitoring, problem resolution tooling and are “VMware-aware”.
•
•
•
V I S I O NS TR AT E G Y
K E Y CO N S I D E R AT I O N SPA S S D O W N
Service Mgt Area
Virtualization assessment criteria (high-level) DatacenterVP/Director Virtualization Center of Excellence
Ch
ang
e M
anag
emen
t
What information is required for changes to virtual machines and how is this recorded?
What constitutes a virtual infrastructure change, and what categories of changes are there (for example, VM virtual hardware memory)?
Are processes documented?
Are different procedures followed for the assessment and approval of normal or complex changes as opposed to simple changes (for example, migration of a virtual machine across a farm group)?
Con
fig
urat
ion
Man
agem
ent What are the purpose, scope, and objectives
of configuration management in a virtual infrastructure?
What are the SlAs for the virtual machine systems?
Are permissions used to limit manipulation of the virtual infrastructure?
How are virtual machines tracked in the Configuration Management Database (CMDB)?
How are virtual machine versions mapped to patch levels?
Inci
den
t M
anag
emen
t How are incidents identified as virtual infrastructure related?
How are virtual infrastructure related issues captured in a knowledge base?
How are virtual infrastructure components monitored?
Prob
lem
M
anag
emen
t
Are there separate procedures to isolate problems related to the virtual infrastructure?
How does the way that virtual infrastructure problems are addressed compare to other platforms?
How are known virtual infrastructure errors logged?
Rele
ase
Man
agem
ent
Are standard build blueprints used within the virtual infrastructure?
Are virtual infrastructure releases tested prior to implementation?
Is a back-out plan developed for each virtual infrastructure release?
Are the master copies of all software in use within standard builds stored in a single repository (Definitive Software library)?
Serv
ice
hel
p D
esk
What tools are used to record virtual infrastruc-ture related issues and how are they classified?
How are escalations of virtual infrastructure related issues escalated?
P h A S E 4: m A N AG E m E N T
45
Service Mgt Area
Virtualization assessment criteria (high-level) DatacenterVP/Director Virtualization Center of Excellence
Ch
ang
e M
anag
emen
t
What information is required for changes to virtual machines and how is this recorded?
What constitutes a virtual infrastructure change, and what categories of changes are there (for example, VM virtual hardware memory)?
Are processes documented?
Are different procedures followed for the assessment and approval of normal or complex changes as opposed to simple changes (for example, migration of a virtual machine across a farm group)?
Con
fig
urat
ion
Man
agem
ent What are the purpose, scope, and objectives
of configuration management in a virtual infrastructure?
What are the SlAs for the virtual machine systems?
Are permissions used to limit manipulation of the virtual infrastructure?
How are virtual machines tracked in the Configuration Management Database (CMDB)?
How are virtual machine versions mapped to patch levels?
Inci
den
t M
anag
emen
t How are incidents identified as virtual infrastructure related?
How are virtual infrastructure related issues captured in a knowledge base?
How are virtual infrastructure components monitored?
Prob
lem
M
anag
emen
t
Are there separate procedures to isolate problems related to the virtual infrastructure?
How does the way that virtual infrastructure problems are addressed compare to other platforms?
How are known virtual infrastructure errors logged?
Rele
ase
Man
agem
ent
Are standard build blueprints used within the virtual infrastructure?
Are virtual infrastructure releases tested prior to implementation?
Is a back-out plan developed for each virtual infrastructure release?
Are the master copies of all software in use within standard builds stored in a single repository (Definitive Software library)?
Serv
ice
hel
p D
esk
What tools are used to record virtual infrastruc-ture related issues and how are they classified?
How are escalations of virtual infrastructure related issues escalated?
4�
CREATING THE BASE PLAN WITH THE MAGNETS
• Simply place the calendar magnets horizontally
• Place the deployment scope and strategy magnets vertically as shown
• Then place the VI processes optimized, critical learning and return on
investment timeframe markers as per your needs over time.
INSTRUCTIONS: PHASED DEPLOYMENT PLANNING CALENDAR
As you rollout your virtual infrastructure operations vision and strategy
and you socialize the tactics and scope of VI deployments with
various internal teams, (for example selecting tier one and tier two
work loads to vir tualize) you will undoubtedly use a whiteboard to
brainstorm. We have enclosed some vir tual infrastructure magnets
for use on your whiteboard.
Use your dry erase markers as usual and these magnets to i l lustrate
ideas. This is an effective way to iterate planning and to depict your
deployment with your team.
O V E R V I E W: V I R T UA L I N F R A S T R U C T U R E P h A S E D D E P LOY m E N T P L A N N I N G C A L E N D A R
Calendar Components
VI Deployment Scope
VI Deployment Strategies
R E S O U R C E S
VI Processes Optimized