Federal Data Center Consolidation Initiative
May 12, 2010
1
FDCCI Business Drivers
 The reported number of Federal data centers grew from 432 in 1998 to more than
1,100 data centers in 2009[1].
 This growth in redundant infrastructure investments is costly, inefficient, unsustainable
and has a significant impact on energy consumption.
 In 2006, Federal servers and data centers consumed over 6 billion kWh of electricity
and without a fundamental shift in how we deploy technology it could exceed 12 billion
kWh by 2011[2].
 In addition to the energy impact, information collected from agencies in 2009[1] shows
relatively low utilization rates of current infrastructure and limited reuse of data centers
within or across agencies.
 The cost of operating a single data center is significant, from hardware and software
costs to real estate and cooling costs.
[1] Current Data Center count is based on OMB Budget Data Request (BDR) No. 09-41, August 8, 2009
[2] Report to Congress on Server and Data Center Energy Efficiency, Public Law 109-431, U.S. EPA ENERGY STAR Program, August 2, 2007
2
FDCCI Strategic Goals & Primary Benefits
The Federal Data Center Consolidation Initiative aims to address these challenges
by leveraging best practices in the public and private sector, and focusing to:
1. Promote the use of Green IT by reducing the overall energy and real estate footprint of
government data centers;
2. Reduce the cost of data center hardware, software and operations;
3. Increase the overall IT security posture of the government; and
4. Shift IT investments to more efficient computing platforms and technologies.
Primary Benefits:
 The primary benefit from improved IT equipment utilization is reduced overall
energy consumption which leads to significant energy cost savings (up to 90%
for virtualized IT equipment)
 Additional benefits from Virtualization & Data Center Consolidation
•
•
•
Reduced Facilities Maintenance & Operations costs
Reduced Server Maintenance & Operations costs
Improved Automation for Server Management & Provisioning
3
Data Center Consolidation – Utilization Improvement Metrics
Improving IT asset utilization is the key driver for reducing energy
consumption per unit of performance. This can be achieved primarily by:
– Server Virtualization (increasing the number of virtual servers per hosts)
– Server Consolidation (decommissioning underutilized physical servers)
– Rack Space Consolidation (relocating underutilized racks)
– Data Center Consolidation (shutting down underutilized facilities)
Utilization Metrics
Typical Results
Target Results
Average Virtualization (%)
0-10%
30-40%
Average Virtual OS per Host (#)
5-10
15-20
Average Server Utilization (%)
7 – 15%
60 – 70%
(application dependent)
Average Rack Space Utilization (%)
50 – 60 %
80 – 90%
Average Power Density Usage
Equivalent (W/sq.ft.)
50 – 100 W/Sq Ft
150 – 250 W/Sq Ft
Power Usage Efficiency (PUE)
3–2
1.6 – 1.3
4
High-Level Summary of Consolidation Approaches
Approach
Decommission
Centralization /
Site Consolidation
Description
Potential Benefits
Rationale
Turn off servers that are not
being used or used infrequently (e.g. dedicated
development environments)
• Cost Savings
• Energy Efficiency
• Frees Floor / Rack Space
• As many as 10-15% of servers
may be inactive but still powered
on in data centers*
Move servers/storage to a
few selected data centers
•
•
•
•
Floor Space Cost Savings
Operational Cost Savings
Increase Rack Utilization
Energy Efficiency
• Approximately 430 Government
data centers are categorized as
“closets” or small sized data centers
(less than 1,000)**
• Server Utilization is approximately
21% Government wide**
Consolidate small data
centers to larger target
centers
Virtualization
Consolidate several
servers onto a single server
through virtualization of the
OS/Platform
•
•
•
•
Floor Space Cost Savings
Increase Rack Utilization
Increase Server Utilization
Energy Efficiency
Cloud
Computing
Alternatives
Move application functions
to standard, vendor
supported enterprise
platforms or services
•
•
•
•
•
•
Floor Space Cost Savings
• Reduce Operational Risk, lower
Energy Efficiency
TCO and TCSD
Operational Cost Savings
• Approximately 40% of Civilian
Cap Ex Cost Savings HW/SW Agency Systems are low-impact
Reduced SW Maintenance
FISMA security, and therefore may
Improved Service Delivery
be low-risk candidates for Cloud
Computing solutions
* McKinsey Report: Revolutionizing Data Center Efficiency, July 2008
** OMB BDR 09-41 Data Analysis, October, 2009
5
Data Center Consolidation – Agency Reporting Schedule
Deliverables
1. INITIAL ASSET
INVENTORY
Description
Conduct an initial inventory of data center assets. This will provide a highlevel understanding of the scale and size of data centers, IT infrastructure
assets, and supported applications.
Agency
Deadlines
April 30, 2010
Develop an initial data center consolidation plan. This plan will identify
2. INITIAL DATA CENTER potential areas for consolidation, areas where optimization through server
CONSOLIDATION PLAN virtualization or cloud computing alternatives may be used, and a high-level
transitioning roadmap.
June 30, 2010
Collect the final asset inventory baseline containing more detailed data.
3. FINAL ASSET
This will serve as the foundation for developing the final data center
INVENTORY BASELINE
consolidation plans.
July 30, 2010
4. FINAL DATA CENTER
CONSOLIDATION
PLANS
Develop final data center consolidation plans. The final plans will include a
technical roadmap and approach for achieving the targets for infrastructure
utilization, rack density and consolidation.
Aug. 30, 2010
5. ONGOING
MONITORING
Conduct ongoing annual monitoring and reporting staring in FY11:
•Update the asset inventory annually (end FYQ3)
•Report progress on executing data center consolidation plans (end FYQ4)
•Reflect changes in asset inventories and account for execution of data center
consolidation plans in subsequent year’s budget
June 30, 2011
Sept. 30, 2011
6
Strategic Goals and Opportunities
Strategic Goals
Tactical Opportunities
Software Assets
Hardware Assets
Facilities & Energy
Location & Real Estate
Identify/Classify Business Application Services
Reduce the
Cost of IT
Lower Leasing Costs per Sq Foot
Move Applications / Infrastructure to Cloud
Perform Virtualization of Servers / Storage
Consolidate Small Data Centers
Optimize Large Data Centers
Promote the Use
of Green IT
Utilize Alternative “Green” Energy Sources
Optimize Cooling Methods / Performance
Power Supply Improvements
Increase Security
Posture
Invest in Efficient
Platforms &
Technologies
Improve Security via HW & SW Asset and Configuration Management
Improve Performance via Redundancy, Load Balancing, COOP
Aggregate Administrative Functions
Disseminate Government Best Practices
Facilitate Cross-Government Collaboration
Create Metrics for Measuring Goals
7
Primary Focus Areas, Consolidation Approaches and
Constraints/Considerations
FOCUS AREAS
Data Center
Asset
Key Metric
Servers &
Mainframes
Storage
(physical &
network)
• Average Server
Utilization (%)
• Average Storage
Utilization (%)
CONSOLIDATION
APPROACHES
CONSTRAINTS /
CONSIDERATIONS
Decommissioning
Facilities Lease & Term
Agreements
Centralization / Site
Consolidation
App Requirements
(Performance, Security,
COOP/DR, Integration)
• Average Rack
Utilization (%)
Virtualization
Data Center
Floor
Space
• Total IT Floor
Space (sq ft)
Cloud Computing
Alternatives
IT Energy
Efficiency
• Energy
Efficiency (PUE
and kW)
Racks
OS / Platform &
Standardization
Tier Classification,
Square Footage
Available
Geographic Location
(Cost/sq ft, Geo-diversity,
Energy Costs)
Personnel Relocation,
Costs & Risks
8