
Improved Efficiency & Reliability for Servers Using Immersion Cooling Technology
Data centers have traditionally been high power consumers. In a "green" data center, instead of being air-cooled, servers are submerged in a non-toxic dielectric fluid that is 1200 times more effective at whisking away heat. By using state of the art liquid immersion cooling, the power required to run each server is reduced by up to 50% and the power required to cool a server room is reduced by up to 90%. For perspective, power costs represent from 25-40% of monthly data center operating expenses. Any energy expended other than that to drive the IT Load contributes to inefficiency.
Immersion technology largely eliminates the temperature swings that can lead to processor and electronics failures. Micro-arcing is also eliminated as a concern since there is no dust or oxygen in the oil environment. There is also a potential to further mitigate against tin whisker failures since the bath could break whiskers or coat them to prevent shorts. Risk of failure due to airborne contaminants like dust, debris, and corrosive gases is eliminated. The improved heat dissipation of immersion technology also allows comfortable operation of cutting-edge, higher powered servers. So, an immersion cooled data center draws much less power and consequently has a smaller carbon footprint than conventional data centers.
Preparing a server for immersion is fairly straightforward. Simply remove the fans, replace any thermal paste with foil, and encapsulate the boot disk or use a solid state drive ( SSD). Working on a server is equally straightforward. Pull the server out of the fluid and wait for it to drain. Keep a nearby paper towel for any drips.
Learn more about this interesting technology by registering today.
What | Improved Efficiency & Reliability for Servers Using Immersion Cooling Technology |
Presenter | Cheryl Tulkoff, DfR Solutions |
When | Thursday, the 20th of November 2014 |
Where | Online |
Duration | 1,5 hour |
Time | 5:00 PM - 6:30 PM CET - Register |
Time | 8:00 PM - 9:30 PM CET - Register |
This webinar presents results regarding several package types from tin-silver-copper (SAC305) solder to SnPb eutectic solder. It is important to understand the reliability risks associated with the reballing procedure, particularly as it relates to thermal cycling, shock and vibration environments.
Root Cause Analysis of HALT FailuresHALT can be an important step in best practice reliability activities but its use is extremely limited without root-cause analysis. In HALT, failures are frequently treated in either a "fix and move on" or "ignore as not relevant" fashion. As a result, organizations experience significant losses from unreliable equipment and processes as they repeatedly correct rather than prevent problems. This is a reactive rather than a proactive method of solving, managing an eliminating undesirable events.