evlbi tests
- Contemporary messages sorted: [ by date ] [ by thread ] [ by subject ] [ by author ] [ by messages with attachments ]
From: <Chris.Phillips_at_email.protected>
Date: Wed, 20 Jun 2007 16:38:07 +1000 (EST)
Hi All,
Just thought you would be interested in knowing the results of the evlbi
testing Adam and I did today.
We were able to get Mp-Pa-At working at 256 Mbps streaming into the CPSR2
cluster in a similar way to March. This ran in a stable manner for a
couple of hours. Some tests indicate that the CSPR2 cluster should also be
able to sustain 512 Mbps data rates.
We also included Hobart for a few hours. The system could run at 64 Mbps
without any dropouts for an hour in the middle of the day (just - I doubt
it would run for 12 hours with no problems).
However we then ran with a modified version of the recorder software which
drops 1 seconds of data when more than 3 seconds of data is buffered. In
this mode we ran at 128 Mbps on all telescopes for about an hour with no
problems. Every couple of minutes a couple of seconds of data is dropped.
This needs more testing but seems it may be a viable operation mode. Once
the new front end for CSPR2 is installed we should be able to run with a
mixed mode setup reliably (Hobart at 128 Mbps, Pks-Mop-ATCA at 512 Mbps).
Cheers
Chris
Received on 2007-06-20 16:38:24
Date: Wed, 20 Jun 2007 16:38:07 +1000 (EST)
Hi All,
Just thought you would be interested in knowing the results of the evlbi
testing Adam and I did today.
We were able to get Mp-Pa-At working at 256 Mbps streaming into the CPSR2
cluster in a similar way to March. This ran in a stable manner for a
couple of hours. Some tests indicate that the CSPR2 cluster should also be
able to sustain 512 Mbps data rates.
We also included Hobart for a few hours. The system could run at 64 Mbps
without any dropouts for an hour in the middle of the day (just - I doubt
it would run for 12 hours with no problems).
However we then ran with a modified version of the recorder software which
drops 1 seconds of data when more than 3 seconds of data is buffered. In
this mode we ran at 128 Mbps on all telescopes for about an hour with no
problems. Every couple of minutes a couple of seconds of data is dropped.
This needs more testing but seems it may be a viable operation mode. Once
the new front end for CSPR2 is installed we should be able to run with a
mixed mode setup reliably (Hobart at 128 Mbps, Pks-Mop-ATCA at 512 Mbps).
Cheers
Chris
Received on 2007-06-20 16:38:24