[Halld-online] [New Logentry] crashless gluon run

[davidl at jlab.org]

Logentry Text:
--
We ran the DAQ over the weekend using 2 Data Concentrators running on gluon53 (AMD opteron) and gluon54 (Intel Xeon, see specs below). The run was started on Friday afternoon and ran continuously until Monday morning. It ran at 4kHz acquiring 800M events. The configuration used was "hd_fdc_mc.tss" which used the Trigger Supervisor and its built-in internal random trigger. The events were from simulated data, preloaded onto the ROCs. This configuration was used on Friday and shown to crash a two DC system when running them on our Ivy Bridge EMU nodes. Some specs for the CPUs that were able to successfully run as Data Concentrators without crashing are:

gluon53
----------
AMD Opteron 6380  @ 2.5GHz
2 CPUS = 32 real cores
RAM: 64GB
Motherboard: Supermicro

gluon54
----------
Intel Xeon E5645 @ 2.4GHz (Westmere)
2 CPUs = 12 real cores (+12 HT)
RAM: 200GB  (not a typo!)
Motherboard: Dell


This supports the existing evidence that required components for crashing are:
Hardware: Supermicro motherboard with Ivy Bridge or Sandy Bridge CPU
Software: CODA DC or PEB



---

This is a plain text email for clients that cannot display HTML.  The full logentry can be found online at https://logbooks.jlab.org/entry/3291039
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailman.jlab.org/pipermail/halld-online/attachments/20140811/92a9d922/attachment.html