HP ProLiant DL140 Dynamic Power Capping TCO and Best Practices White Paper (WW - Page 2

Executive Summary, Introduction, The Dilemma of Trapped Power Capacity - power consumption

Page 2 highlights

Executive Summary Today, more than ever, companies must focus on optimizing their usage of power and cooling resources. The capital cost of constructing data center infrastructure as well as the ongoing expense of powering IT equipment have risen significantly over the past decade. Data center managers also may encounter limitations, imposed by local utilities, on the amount of power capacity available for IT operations. Gone are the days when the IT staff could assume that they had the power and cooling they required to meet the growing demand for IT services. The Hewlett-Packard Thermal Logic portfolio helps customers reduce data center power consumption, reclaim a large portion of trapped data center power and cooling capacity, and extend the life of existing data center facilities. This paper will focus on an important new component of the Thermal Logic portfolio called Dynamic Power Capping. Delivered only with HP ProLiant and HP BladeSystem, Dynamic Power Capping helps customers reclaim trapped power capacity by safely limiting peak power consumption without impacting system performance. This paper explores how power allocation typically works in the data center, outlines expected Dynamic Power Capping total cost of ownership benefits and outlines best practices for Dynamic Power Capping implementation. Introduction HP offers Dynamic Power Capping to help customers reclaim stranded power in existing data centers or to optimize use of power and cooling capacity in new data centers. Dynamic Power Capping allows the data center manager to cap power, limiting peak power consumption without impacting performance and without risk of over-subscribing data center branch circuits This paper: • Describes how Dynamic Power Capping functions and helps reclaim trapped power and cooling capacity. • Examines the TCO benefits that can be expected when implementing Dynamic Power Capping • Lays out the best approach to implementing Dynamic Power Capping The Dilemma of Trapped Power Capacity As information technology continues to evolve from a tool for enhancing workforce productivity to an engine for innovation and competitive advantage, IT organizations find themselves under increasing pressure to deliver IT services with close to 100% uptime. At the same time, the cost of delivering IT infrastructure continues to rise. Today's Tier IV data centers may cost as much as $25 million for a 1 megawatt of power and cooling capacity.1 The requirement to provide continuous access to IT services with almost no downtime, coupled with significant increases in the cost of IT infrastructure delivery presents a dilemma for many IT organizations. On the one hand, an inability to effectively monitor and control power consumption and an emphasis on uptime and performance optimization encourage a conservative data center management culture. Based on the available tools, power and cooling capacity typically is budgeted to account for theoretical worst-case power usage, even though excess capacity may never be used. 1 Cost Model: Dollars per kW plus Dollars per Square Foot of Computer Floor, W. Pitt Turner IV, P.E. with Kenneth G. Brill. 2

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15

Executive Summary
Today, more than ever, companies must focus on optimizing their usage of power and cooling
resources.
The capital cost of constructing data center infrastructure as well as the ongoing expense
of powering IT equipment have risen significantly over the past decade.
Data center managers also
may encounter limitations, imposed by local utilities, on the amount of power capacity available for IT
operations.
Gone are the days when the IT staff could assume that they had the power and cooling
they required to meet the growing demand for IT services.
The Hewlett-Packard Thermal Logic portfolio helps customers reduce data center power consumption,
reclaim a large portion of trapped data center power and cooling capacity, and extend the life of
existing data center facilities.
This paper will focus on an important new component of the Thermal
Logic portfolio called Dynamic Power Capping.
Delivered only with HP ProLiant and HP
BladeSystem, Dynamic Power Capping helps customers reclaim trapped power capacity by safely
limiting peak power consumption without impacting system performance.
This paper explores how
power allocation typically works in the data center, outlines expected Dynamic Power Capping total
cost of ownership benefits and outlines best practices for Dynamic Power Capping implementation.
Introduction
HP offers Dynamic Power Capping to help customers reclaim stranded power in existing data centers
or to optimize use of power and cooling capacity
in new data centers.
Dynamic Power Capping
allows the data center manager to cap power, limiting peak power consumption without impacting
performance and without risk of over-subscribing data center branch circuits
This paper:
Describes how Dynamic Power Capping functions and helps reclaim trapped power and
cooling capacity.
Examines the TCO benefits that can be expected when implementing Dynamic Power
Capping
Lays out the best approach to implementing Dynamic Power Capping
The Dilemma of Trapped Power Capacity
As information technology continues to evolve from a tool for enhancing workforce productivity to an
engine for innovation and competitive advantage, IT organizations find themselves under increasing
pressure to deliver IT services with close to 100% uptime.
At the same time, the cost of delivering IT
infrastructure continues to rise. Today’s Tier IV data centers may cost as much as $25 million for a 1
megawatt of power and cooling capacity.
1
The requirement to provide continuous access to IT services with almost no downtime, coupled with
significant increases in the cost of IT infrastructure delivery presents a dilemma for many IT
organizations.
On the one hand, an inability to effectively monitor and control power consumption
and an emphasis on uptime and performance optimization encourage a conservative data center
management culture.
Based on the available tools, power and cooling capacity typically is budgeted
to account for theoretical worst-case power usage, even though excess capacity may never be used.
1
Cost Model: Dollars per kW plus Dollars per Square Foot of Computer Floor, W. Pitt Turner IV, P.E. with Kenneth G. Brill.
2