Managing Data Center Consolidation to Optimize Efficiency and Cost Savings Peter A. Panfil VP and General Manager Liebert AC Power Emerson Network Power
Agenda • • • •
Federal data centers to stop infrastructure sprawl Key considerations in the changing data center infrastructure Consolidation optimizes infrastructure and minimizes footprint Assessing power and cooling capabilities to support consolidation • Monitoring and infrastructure management enhance visibility and control • Emerson’s proven solutions • Q&A
2
Federal data centers to stop infrastructure sprawl
3
Federal consolidation initiatives Reduce energy consumption by 30% by 2015 (Energy Independence and Security Act) Tasked with measuring greenhouse gas emissions and setting targets for reducing them
Four deadlines in six months
Government spends $19 billion on infrastructure
Energy consumption is expected to double by 2011 Agencies use only 15-20 percent of capacity
Source: Federal News Radio article published March 1, 2010, “OMB taking a deeper look at data centers.�
4
Federal Data Center Consolidation Initiative
Initial Data Center Asset Inventory
Final Asset Inventory Baseline
On-going Inventory and Metrics
Source: Federal News Radio article published March 1, 2010, “OMB taking a deeper look at data centers.�
5
Top agency concerns and drivers • • • • • •
IT security and cyber-security Standardization and consolidation: bring efficiency, cut complexity , cut costs Sustainability, Energy Efficiency IT governance – Federal CIO TechStat initiative Transparency Achieving outcomes based results
• Secure and trusted information sharing • Integrated departmental governance • Comprehensive cyber-security initiative • Modernizing legacy systems • Case load/rising healthcare costs • Ensuring cost-effective IT investments to support VA’s mission and mitigating risks associated with IT • Managing and protecting information systems to ensure information availability, integrity, authentication, and confidentiality • Addressing weakness in IT security controls • Achieve a persistent and dominant information advantage for ourselves and mission partners/ • Change policies, processes, and culture necessary to provide the speed, accuracy, and agility to ensure mission success in a rapidly changing and uncertain world. • Ensure a robust and secure information environment• • Provide modern command and control capabilities through persistent collaboration at all levels and among all mission partners. • Acquire new information capabilities rapidly (9-12 months) and at low cost by delivering them as enterprise services. • Innovation through technology • Protection of Information Resources and Critical Infrastructure
6
Where federal agencies are going
Consolidate and reduce within five years
Policy against expanding beyond current levels
Consolidate footprints that span several states
Data center consolidation by 2012 to save $3 billion Source: Federal News Radio article published March 1, 2010, “OMB taking a deeper look at data centers.�
7
Key considerations in the changing data center infrastructure
8
Powerful forces driving change in the data center infrastructure COMPLEXITY IS GROWING AND RATE OF CHANGE IS INCREASING External forces changing the business climate
Virtualization, cloud computing Interdependent applications/ functions
Data center and server consolidation
Efficiency and green initiatives
Do more with less
Business and technology forces pressing on the data center
Higher density
Collaborative IT and Facilities planning
Higher end user demand
9
New UPS systems provide optional improvements • New intelligent UPS controls have been developed to deliver higher system efficiency without sacrificing reliability • UPS Systems now running in Eco-Mode with line power conditioning Data Center B Energy Data Center A Energy Savings Per "Efficient" Year 360 kW of IT, 10 cents / kW hr UPS effeciency 94.5% 97.5% Cost to power the load $298,015 $307,476 $9,461
• Intelligent Paralleling • Automatically idles un-needed UPS equipment • Ensures proper redundancy is maintained
10
Better to equate efficiency to lifecycle of a data center Ability to produce an output with a minimum of effort, expense, or waste
Design & Deployment
Management & Planning
Operation
11
Optimizing data center infrastructure to reduce cost and deliver high availability
12
Consolidation optimizes infrastructure and minimizes footprint
13
Drivers behind consolidation • Improved server utilization • Reduced number of servers ‒ Consolidation ratios of 8:1 are typical • Energy efficiency ‒ Potential savings of up to 14% on energy costs • Increased data center flexibility to meet changing business needs
14
Key consolidation considerations • Consolidation Reduces
• Consolidation Increases
– Power usage
‒ Performance
– Heat – overall
‒ Utilization
– Wiring – when managed properly
‒ Heat – in a given rack
– Rack space required
‒ Power consumption – in a given rack ‒ Wiring management challenges
15
Virtualization has created a dynamic applications environment Previous approach to Provisioning App1
App2
App3
App4
Today’s approach to Provisioning App 1
App 2
App 3
App 4
App 5
App 6
Virtualization IT Infra
IT Infra
IT Infra
Physical Infrastructure
IT Infra
IT Infra
IT Infra
IT Infra
Physical Infrastructure
With Virtualization, physical infrastructure requirements can change as load shifts
16
Exposing a critical gap in data center operations • No tools today to synchronize the virtualization automation with the physical layer • Results in potential overloading and subsequent outages • Does not allow optimization in design There is a need for physical and IT infrastructure to have a dynamic relationship
App 9
App 1
App 2
App 3
App 4
App 5
App 10
App 6
App 11
App 7
App 12
App 13
App 8
Virtualization IT Infra
IT Infra
IT Infra
IT Infra
X X Physical Infrastructure 17
Power and cooling can improve success • Total efficiency benefits of consolidation and virtualization may not be realized if power and cooling strategies are not adjusted • New power and cooling strategies can improve results – Add more servers to each rack – Increase server availability and utilization – Optimize data center reliability and efficiency – Enhance flexibility for future growth – Reduce energy consumption
18
Consolidation increases criticality “With virtualization, since each server runs multiple virtual server workloads, the server becomes a single point of failure. This escalates the issue of availability for that physical server. If you put all your virtual eggs in one basket, then you need to take very good care of that basket.�
- Barb Goldworm, Blade Servers and Virtualization
19
Assessing power and cooling capabilities to support consolidation
20
Assessing power and cooling needs • A data center assessment or walkthrough will help to identify, evaluate and resolve performance and energy efficiency issues – Airflow assessment – Branch circuit power – Grounding – Infrared inspections • Ideal for beginning of a project or for planned or existing virtualized environment
21
Power management goals • Power infrastructure adapts with changing requirements for capacity, availability and redundancy • Infrastructure protects critical applications that must stay operational
22
Redundancy and availability: Eliminate single points of failure • Redundant UPS • Redundant PDU • Redundant circuit
U P S
Room Level
Rack Level
U P S
UPS PDU
Power Dist. Panel
PDU UPS
Power Dist. Panel
Utility Outlet
Utility Outlet
23
Accommodate future growth • Add rack UPSs if the data center has just a few racks OR • Once power reaches 15 – 20 kW, use room-level UPSs – Software scalability increases capacity with a software key without adding hardware • 40kVA → 60kVA → 80kVA • 80kVA → 100kVA → 120kVA
24
Cooling management goals • Cooling adapts with computing demand • More effective cooling of high-density hot spots • Optimized data center space & efficiency – Reduces energy costs
25
Dynamic precision cooling • Multiple densities in the same room and same row • Heat issues at the rack level • Digital controls allow more precise tracking of cooling to achieve both availability and efficiency • New strategies enhance energy efficiency
26
Cold aisle containment • Cooling with existing underfloor air • Focused cooling increases capacity and efficiency • Increases energy efficiency by up to 33% • Typically supports 10 - 12kW per rack • Enables retrofit without disrupting operations
27
Row–level cooling •
•
•
• •
As densities increase, it becomes inefficient to move large volumes of air from edge of room to the heat source and back Row-level cooling can – Bring cooling closer to heat source – Reduce amount of energy required to circulate air – Increase capacity and efficiency of the cooling system Localized cooling requires less floor space Row-level cooling can reduce operating and energy costs May qualify for utility rebates 28
High density cooling designs • High-density cooling moves cooling closer to the heat source • Greatly reduces number of room cooling systems – saves floor space • Adapts to change in the data center – easy to add cooling modules • Saves energy – Can increase energy savings by up to 18%
29
Efficiency benefits of high density Traditional Cooling Only Fan Power - 8.5kW per 100 kW of cooling Average entering air temperature of 77 F Cooling Unit
Supplemental and Base Cooling Fan Power- 3.0 kW per 100 kW of Cooling (Liebert XD @ 2 kW per 100kW)
Average entering air temperature of 98 F
• 65% less fan power • Greater cooling coil effectiveness • 100% sensible cooling
30
Visibility into data center operations allows better control • Virtualization creates a more complex environment • Increased visibility and control is needed because server utilization changes dramatically throughout the day • Real-time access to temperature data, utilization rate, power and cooling system data combined with historical time of day/seasonal models
31
Monitoring and infrastructure management enhance visibility and control
32
Data collection is critical to infrastructure management • • • • •
• • • •
UPS Battery system Cooling Leak detection Backup generator Surge protection Switchgear Pumps and chillers Power distribution / branch circuit
33
“Traditional� data center too limiting: Thinking focused on functional layers
IT and Physical Infrastructure was designed for STATIC Application requirements 34
Bridging facilities and IT management
? Companies lack cohesive management strategy for data center infrastructure 35
Data center infrastructure management maturity MONITOR AND ACCESS
• How are my assets operating? • Am I getting realtime notification of alarms and alerts? • How do I get my server back up and running? • Can I populate my planning tools with actual performance data?
Early Warning (Reactive)
DATA CAPTURE AND PLANNING
ANALYZE, DIAGNOSE
RECOMMEND AND AUTOMATE
• What and where are assets in the data center?
• How do I extend the life of the data center?
• How are they interconnected?
• How do I reduce mean time to repair (MTTR)?
• How do I anticipate potential failures and automatically shift compute and physical load to eliminate downtime?
• Do we have space, cooling and power to meet future needs? • How can I efficiently commission decommission?
Improved Planning
• How do I synch infrastructure with virtualization automation? • How are we doing against SLAs?
Reduced MTTR and Effort
• How can I optimize efficiency across my data center?
Availability at Optimal Performance (Proactive)
Customers need to evolve through levels of maturity in DCIM
36
Energy Star Program Server Operating Efficiency
• More than 30 percent of servers in the U.S. operate at less than 2 percent capacity • Roughly half of the energy consumed by data centers is used to cool the center, not to power the IT equipment • Most data centers operate at 70 degrees; however, today’s servers can handle temperatures above 90 degrees
Guiding Principles
Energy Star Rating
• PUE is the preferred energy efficiency metric • IT energy consumption should be measured at the output of the UPS • Energy Measurement specifications: • Dedicated – point of utility handoff • Mixed-use – all energy required to operate the facility
• Getting started – strategic energy management is key • Control energy consumption and operating costs • Benchmark your data center – access Portfolio Manager • Receive energy performance rating • Earn Recognition – be honored by EPA for energy efficiency initiatives
Product Specifications
• Computer Server Specification Development • Data Center Storage Specification Development • Uninterruptible Power Supplies Specification Development
Source: Energy Star Data Center Energy Efficiency Initiatives at www.energystar.gov/index.cfm?c=prod_development.server_efficiency 37
Infrastructure solutions leverage industry leading expertise Automatic Transfer Switch
Paralleling Switchgear
Fire Pump Controller
Surge Protection
Uninterruptible Power Supplies & Batteries
Integrated Racks Precision Cooling
Cold Aisle Containment
Cooling
Rack
Power Distribution Unit
Extreme-Density Precision Cooling
KVM Switch
Infrastructure Management & Monitoring
UPS Power Distribution Units
Monitoring Server power supplies
38
Emerson Network Power solution overview Infrastructure Monitoring & Management
Access & Control
Power Management
Precision Cooling
Rack Solutions
Modeling, planning, tracking and complexity reduction
Suite of market leading solutions
Best practices in Data Center Infrastructure Deployment and Management
Service Capabilities • Industry Leadership: Energy Logic, Emerson Green Data Center • Best practices assessments and implementation service
Ability to simplify vendor management
Domain expertise across critical data center infrastructure
Best Practices
Visibility, Access & Contol at the rack & device
Embedded Technology Solutions
Service Processor Manager
KVM Switch
Sensor Network
Serial Console
Rack Distribution Unit
PDU
Flexible Planning and Design of Infrastructure components
39
Emerson’s proven solutions
40
Emerson’s innovative solutions focus on data center optimization MONITOR AND ACCESS
DATA CAPTURE AND PLANNING
DIAGNOSIS AND INSIGHT
Assessments, Benchmarking Liebert Nform, Liebert SiteScan, Avocent DSView Services
Aperture & Avocent Infrastructure Explorer Services
Aperture IRM Services
Improved Planning
Reduced MTTR and Effort
RECOMMEND AND AUTOMATE
Ongoing Energy Efficiency
Emerson Vision
Availability at Optimal Performance
Energy Optimization
Earlier Warning
Emerson has products today and continues to invest in solutions that close the gap and help you evolve your management of heterogeneous infrastructure 41
Holistic solutions
42
Energy Logic: Prioritized energy saving strategies
Higher AC voltage improves efficiency
43 Emerson Electric Co.; Proprietary Information
Unoptimized data center layout
44
Optimized data center layout
45
Liebert customer portfolio • • • •
DoD Civilian Agencies National Labs Intelligence Agencies
• Prime Contractors • Lockheed Martin • SAIC • Siemens
46
Real results with Liebert solutions Sun’s Santa Clara Data Center • Sun performed steps 1-5 • Liebert performed steps 6-10 Results: • Over 60% reduction in energy costs • Reduced Sun’s carbon emissions by 1% (4,100 tons/yr)
• Awarded over $1 million rebates and awards from utility provider • Mix of redundancy and power requirements
© 2007 Emerson Network Power
47
Efficient Data Center: The POD Concept •
A group of Knurr racks or benches with a common hot or cold aisle used as a building block to simplify data center design for Liebert power, cooling, & cabling.
Slab or Raised Floor, Flexible, Scalable, High Density Š 2007 Emerson Network Power
48
Sun Santa Clara data center efficiency
49
Q&A
Peter A. Panfil VP and General Manager Liebert AC Power Emerson Network Power Peter.Panfil@emerson.com 740.833.8665
50