Monday 2008-02-04

30 nodes in rack down at about 16:00 UTC

All computers and switches were off by about 16:11 UTC

Tuesday 2008-02-05

Computers were started at these times (all UTC):

• 09:10 Switches (tiny fans)
• 09:13 SUN X4500 (10 large fans - acoustically loud)
• 09:15 single server (6 small fans @ ~ 13500rpm)
• 09:16 Aux servers (7 small + 2 file servers, many small fans)
• 09:19 30 compute nodes + 1 head node (close to 200 fans @ ~13500 rpm)

Wednesday 2008-02-06

Thursday 2008-02-07

Planned: Switching on the LCP and outdoor cooling units partially

• 8:40UTC Racks on
• 16:00UTC Racks off

between ~ 07:00UTC to 12:00UTC office containers were put into place.

Friday 2008-02-08

• start of 72 hour test at about 7:20UTC
• at about 08:00UTC on outside chiller switched itself off and remaining 2 are running at 100%
• at around 19:00UTC the power usage was reduced to calm down the chillers

Saturday 2008-02-09

• back to full power at around 05:00UTC-06:00UTC
• during the day: works on the external system, change to a different modus operandi
• Around 23:10UTC complete shut down of system due to failure of control unit
• back to full power between 23:30UTC and 24:00UTC

Sunday 2008-02-10

• full power all day

Monday 2008-02-11

• some tests including reducing the power usage and power failures.

%IMAGE{"00TotalPower.png"}%

%IMAGE{"00AvgWaterIn.png"}%