See also

• 1) CCB High Performance Computers (HPC)

UPS inventory for CCB Clusters and non-cluster HPCs

Note 1: Currently none of the clusters ChemIT manages have UPSs for their compute nodes. Thus, this is our standard community standard of practice. (Is this what CAC does, too?).

• ChemIT staff must be called in to restart the compute nodes after //any// power failure.
• After even the briefest of power outages, all compute nodes will be off and thus clusters will be unusable. This is true even if the headnode and network switch have UPS backup.

Note 2: Having a UPS is expected to provide power backup for perhaps less than 10 minutes. (Depends on size of UPS, condition/ age of that UPS's battery, and the demands placed on that UPS.) This protects against most power outages.

• Adding a USB connection from the UPS to a system allows that system to execute a shutdown command, properly shutting the system down in a timely manner during a prolonged power outage. Otherwise the system will simply lose power, and that kind of forced shutdown can often cause software and hardware failures.
• Ideally, ChemIT will have the resources to expand our capabilities to enable shutting down UPS-protected systems during prolonged power outages running beyond the capacities of the UPS, beyond the systems with a direct USB connection to the UPS.

Power outage impact on systems with and without UPS

~5-10 minute outage from Sunday, 4/23/207, per Michael Hint's investigations

What does it cost to UPS a research system?

Current goal: Do all head nodes and stand-alone computers in 248 Baker Lab. Started getting done, fall 2014.

Assuming protection for 1-3 minutes MAXIMUM:

About $180 (APC brand) per head node or server every ~3-4 years  (3 yr for warranty and ~4 years actual battery life).

• Unusual case: ~$900/ set of 4 GPU systems (Scheraga) every ~3-4 years (for this, must confirm approach, appropriateness, and estimates).

Remaining UPS's to invest in

Clusters

Most we been done Spring 14, after the spate of power failures. See CCB's HPC page (first chart, in "UPS for headnode" column) for details

Non-clusters

See CCB's HPC page (second chart, in "UPS" column) and CCB's non-HPC page (in "UPS" column) for details of the few that are already done.

Stand-alone computers' UPS status:

Review others at above two cited pages which might need a UPS, after above ones done.

Switches

Do all switches: Maybe ~$340 ($170*2), every ~4 years.

• Recommend: Do ASAP.
• Q: Funding?
• Other issues and concerns with actually implementing this approach:
  • Rack space. Maybe power. Maybe cord lengths. What other issues?

Do all compute nodes: ~$18K initially, and perhaps ~$4.5K every ~4 years to replace batteries and deal with UPS hardware failures.

• ~20 20amp UPS's ($900 each) required.
  • Replacement batteries ~$200 each, or ~1/4 replacement cost.
• Estimates are simply back-of-the-envelope calculations.
• If were to actually implement, there may be smarter ways to do this, but the total cost will likely not be lower.
  • In fact, costs may be higher, if sufficiently higher benefit doing it a different way, for example.
• Issues and concerns with actually implementing this approach:
  • Costs. Rack space. Maybe power. Maybe cord lengths. What other issues?

Compute node counts, for UPS pricing estimates. Does not include head node:

• Count source: ChemIT's Computer counts with CCB clusters