15% UPS waste? Really? #
Posted Monday 29th June 2009 17:46 GMT
I'll let others comment on the ridiculous name and make a couple of serious comments:
- If your simulation has been running for 2 months, you really should have been taking snapshots periodically.
- If you were powering your system by charging a battery from the mains and then switching over to power the processor from the battery, then a 15% loss in that process would be impressively good (30% is more likely). But a UPS should not be operating like that: the only loss should be the energy needed to trickle-charge the batteries, which should be tiny. If your UPS is not operating in that way you should fix your UPS, not re-architect your power system around it.


HP shows smallest pod in the world, talks HPC
Exascale computing: How do we get there from here?

