Business Case for IBM System p5 Virtualization Economic

1
Business Case for IBM System p5 Virtualization
Economic Benefits of Infrastructure
Simplification
February 10, 2006
2
This presentation was prepared by the
International Technology Group (ITG) with
International Business Machines Corporation (IBM)
funding. Information for this presentation has
been obtained from sources assumed to be reliable
and reflects conclusions at the time. Although
the presentation may utilize publicly available
material from various vendors, including IBM, it
does not necessarily reflect the positions of
such vendors on the issues addressed in this
presentation. Material contained and conclusions
presented in this presentation are subject to
change without notice. All warranties as to the
accuracy, completeness or adequacy of such
material are disclaimed. There shall be no
liability for errors, omissions or inadequacies
in the material contained in this presentation or
for interpretations thereof. Trademarks included
in this presentation are the property of their
respective owners. This presentation is based
upon information contained in the Business Case
for IBM System p5 Virtualization Economic
Benefits of Infrastructure Simplification
Management Brief dated February 2006, and
released by ITG. This presentation may ONLY be
presented in conjunction with Management Brief.
3
Virtualization Benefits

• Reduced complexity
• Fewer physical servers
• Higher capacity utilization
• Improved manageability
• Service quality (e.g. availability) gains
• Cost savings
• Bottom Line?

4
Focus of Study

• UNIX servers
• Large organization environments
• Broad range of applications workloads
• Potential economic benefits of server deployment
  strategies that effectively exploit the
  virtualization capabilities of the IBM System p5

5
Basis of Calculations (1)Profiles Scenarios

• Conventional scenarios
• These are built around diverse multivendor server
  bases supporting a wide range of applications
  workloads. Bases include different technology
  generations software versions. Conventional
  management operating practices, including
  limited use of partitioning, are employed. These
  scenarios are typical of server environments
  found in large organizations today.
• Virtualized scenarios
• The same applications workloads are deployed on
  System p5 servers, using the full potential of
  System p5 virtualization. Consolidation of
  multiple applications instances onto single
  servers occurs where this is technically
  organizationally feasible. IT management
  operating practices are adapted to effectively
  exploit virtualization potential.

6
Basis of Calculations (2)Server Bases Staffing
7
Types of Costs Compared

• Two types of costs are addressed by the study
• Server costs server hardware, maintenance,
  systems database software, full time equivalent
  (FTE) system administration personnel,
  facilities (data center occupancy, power
  cooling).
• Costs of downtime business losses due to supply
  chain disruption, inability of customers or
  partners to access online systems, other
  effects caused by outages.

8
Server Costs Comparisons(1) Five-year Operating
Costs
9
Areas of Operating Cost Reduction for System p5
Virtualized Scenarios
10
Server Cost Comparisons (2)
Costs for System p5 virtualized scenarios
include initial hardware software acquisition
11
Costs of Downtime

• Costs of downtime are business losses due to
  supply chain disruption, inability of customers
  or partners to access online systems, other
  effects caused by outages. Industry-
  company-specific values are employed in
  calculations.
• Costs of downtime for the financial services
  company are for lost operating profit (net
  revenue after personnel, occupancy, other
  overhead). For the manufacturing retail
  companies, costs are for lost gross profit
  (profit after cost of goods sold).

12
Five-year Comparisons Costs of Downtime
13
Conclusions

• Even if organizations do not undertake
  large-scale, short-term transformations of server
  bases, phased creation of virtualized
  infrastructures over time represents a new,
  highly attractive opportunity to increase IT
  cost-effectiveness.
• Significant opportunity to reduce server
  infrastructure costs results suggest
  opportunities to reduce operating costs to 35
  to 50 of current levels.
• Server costs are only part of the picture.
  Servers support organizational applications
  databases, are one of the central structures
  around which networks are built. Which means that
  the efficiency or otherwise of server
  infrastructures affects both the economics
  effectiveness of the entire corporate IT
  environment.
• Broader potential benefits include
• Higher overall productivity of organizational IT
  expenditure, including application delivery
  business enablement processes
• Faster, more effective response to changing
  business needs
• Improved quality of service across all systems
• Organizations are often wary of quantum leaps
  in technology. But a quantum leap in the
  organization operation of server
  infrastructures has nevertheless become possible.
  Virtualization of servers, as well as of other
  resources, will clearly become a central driver
  of IT change in the years to come.

14
Methodology Configuration Sizing
Business Case for IBM System p5 Virtualization
Economic Benefits of Infrastructure
Simplification Management Brief
In translating configurations employed in
conventional scenarios into System p5
configurations employed in virtualized scenarios,
the following approach was employed. System p5
configurations delivering performance equivalent
to conventional scenario servers were developed
on a case-by-case basis for each of the main
applications in profile companies. Configurations
were developed for production as well as
non-production instances. Application-specific
comparative metrics were used to determine System
p5 processor capacity, as well as memory, I/O and
other resources to support each instance in an
efficient manner while delivering equivalent or
superior levels of service quality. Further
calculations were then undertaken to allow for
the effects of virtualization. Nominal
configuration requirements were first determined
for groups of instances that were deployed on
separate servers, or in HP nPars or vPars, IBM
LPARs or Sun DSDs in conventional scenarios,
where these represented realistic candidates for
consolidation onto a single System p5 physical
server. An overall utilization value reflecting
realistic potential consolidation efficiencies
was then assigned to each group of instances, and
allowance was made for other factors affecting
System p5 capacity requirements. The resulting
configuration was then rounded to next largest
capacity increment offered by IBM e.g., a
nominal configuration of 3.26 x 1.9 GHz CPUs with
10.3 GB of RAM was rounded to a 4 x 1.9 GHz
System p5 model with 12 GB of RAM. Other
hardware components were configured similarly.
Allowance was made for other System p5
virtualization capabilities such as Virtual I/O
and Virtual LANs. In developing virtualized
scenarios, consolidated configurations were based
on realistic commonalities between applications
and instances e.g., database and application
serving, and non-production and production
instances were typically not consolidated onto
the same physical System p5 server. Equally,
production applications controlled by different
business units within companies were in most
cases not consolidated onto the same physical
platform. For each profile, consolidation
occurred only within certain system groups
detailed in the Management Brief. In all of these
areas, virtualized scenario configuration sizing
reflect user best practice experiences.
15
Methodology Server Costs
Business Case for IBM System p5 Virtualization
Economic Benefits of Infrastructure
Simplification Management Brief
Maintenance, software update and support and (for
virtualized scenarios) hardware and software
acquisition costs were calculated based on
discounted vendor United States (U.S.) list
prices. For database software, Oracle 10g dual
core pricing was employed where
appropriate. Systems software stacks for
conventional scenarios include operating systems
and, where appropriate, additional facilities
providing functionality equivalent to that
incorporated in AIX 5L Version 5.3. Systems
software stacks for virtualized scenarios include
AIX 5L Version 5.3 and for clustered
configurations IBM High Availability Cluster
Multi-Processing (HACMP) facilities. Personnel
costs are for numbers of full time equivalent
(FTE) system administrators shown earlier.
Calculations were based on annual average
salaries of U.S. 75,316, increased by 25.3
percent to allow for bonuses, benefits, training
and other items. Facilities costs for
conventional as well as virtualized scenarios
include servers as well as power and cooling
equipment. Server facilities costs were
calculated using vendor specifications for
electricity consumption and footprints for server
models and configurations included in scenarios.
Costs include data center occupancy (calculations
were based on EIA standard rack mount units and
service clearances for these, plus allowance for
inactive areas) and electricity consumption.
Facilities costs also include hardware
acquisition, maintenance, occupancy (including
service clearances and inactive areas) for, and
electricity consumption by power and cooling
equipment. Configurations of this equipment are
appropriate for overall electricity consumption
and heat generation by server bases in scenarios.
Cost calculations were based on U.S.
specifications and discounted list prices for
appropriate models from leading vendors. Data
center occupancy costs for servers as well as
power and cooling equipment are based on a
conservative assumption for annual average cost
per square foot for existing facilities (i.e.,
costs do not include new facilities
construction), while electricity costs for both
are based on a conservative assumption for
average price per kilowatt/hour. Both assumptions
are for U.S. costs.
16
Methodology Costs of Downtime
Business Case for IBM System p5 Virtualization
Economic Benefits of Infrastructure
Simplification Management Brief
Costs of downtime calculations were based on
applications and availability levels detailed in
the Management Brief. For each application and
scenario, the frequency and duration of outages
over a 365-day period was determined based on
user experiences. Industry- and
organization-specific values were then applied to
calculate costs caused by these. Allowance was
made for the nature and timing of outages e.g.,
a planned outage for an online system occurring
between 2 a.m. and 3 a.m. on a Monday morning was
not assigned the same cost value as an unplanned
outage occurring during normal business hours
during a peak month.