Session was very well attended about 20 participants

Data Center greening qualifies for upto 20% Tax credits.
At Cisco about 75% energy usage is in their Labs and Data Centers
NetApp qualified for largest rebate from PGE
EBay, Yahoo have very active projects to lower energy usage

Data Centers are generally considered mission critical so reliability (100% uptime) is essential ... a Tier 4 data center has super redundancy (2 of everything plus 1) so is least efficient.

Must change behavior .. people must learn that instant access to data is expensive and affects environment ..
Must improve software so it runs more efficiently and needs less power

Big companies with multiple data centers can reduce need for redundancy by sharing the load between centers but need better software

One inherent obstacle is that traditionally Facilities is responsible for Buildings and Energy while IT is responsible for computers and service to users.

Hardware design must become more resilient and stand higher temps
Water usage must be considered (in generating power and in cooling the data centers)
E-waste must be considered at end of life server disposal

Energy usage can be lowered by:
Tech refresh practices (replace old servers with newer more efficient ones)
Higher ambient temps (need more data on server performance limits)


Leading practices are not readily disclosed -- deemed to be competitive advantage. EDF could (should?) drive a project to share best practices to avoid imposed regulations and resource scarcity
EBay would be willing to share success story

Metrics must be improved - they are not standardized --
PUE = Power Usage Effectiveness = Ratio of IT Load to Facility Load
CPE = Computer Power Effectiveness
PDU = Power Distribution Unit
DCEP = Data Center Effectiveness ?

Waste heat from data centers could be re-used (one example is in a co-located greenhouse)

UC San Diego has a report has been posted on wiki by John Hailey
Summary below:

Data Center Energy Efficiency The cost of energy consumption in modern data centers has reached and even surpassed the cost of the physical data center itself, necessitating research for dynamically reducing the amount of energy used for computing, cooling and maintaining a data center. The primary goal of this work is the development of a data center power management scheme that delivers energy efficiency with minimal impact on performance. The scheme is based on developing policies for power management techniques like dynamic power management (DPM) and dynamic voltage frequency scaling (DVFS) using online learning and dynamic workload characterization, where the policies adapt to changes in the workloads. Our experiments with CPUs and hard disks confirm the efficiency and adaptability of our online learning algorithms. We further propose extensions to adapt this approach to a virtualized environment encompassing multiple virtual and physical machines, characterizing virtual machines at the hypervisor level to drive both the power management policies and energy aware scheduling. The energy aware scheduler should schedule virtual machines both within and across physical machines for higher energy efficiency.