r1 - 08 Aug 2007 - 10:53:27 - RobertGardnerYou are here: TWiki >  Admins Web > CharlesDataXferIU
Rob:  just for interest, here's a plot of how many jobs were in the
"transferring" state between the end of day yesterday and now.

I collected this data by checking the panda monitor from a shell
script (using lynx) and saving the data to a file

I plotted the file using "gnuplot" and am attaching the output
as a PDF.  Sorry about the x-axis labeling - there are no units
but this is just datapoints, taken about every 10 seconds, between
8PM yesterday (8/6) and 10AM this morning (8/7)

The plot looks like a bunch of upward ramps punctuated by sharp
drop-offs (when FTS decides that a transfer or set of transfers
has succeeded).   The overall slope is downward, but each segment
has upwards slope.

There are still 1649 files in the 'transferring' state at MWT2_IU
but this is down from the high of over 1800 yesterday.... it is
coming down but I don't know if it is coming down fast enough or
if perhaps we still have a problem (eg, slow transfers)

Number of files transferring out of IU_ATLAS_Tier2 is down to 200,
from over 300 yesterday, but again I don't know if this is enough
of a change, if we are clearing the backlog quickly enough.

Our failure percentages our down, according to the PANDA monitor,
MWT2_IU now only has a failure rate of 1%, as compared to 6% at

                - Charles

-- RobertGardner - 08 Aug 2007

About This Site

Please note that this site is a content mirror of the BNL US ATLAS TWiki. To edit the content of this page, click the Edit this page button at the top of the page and log in with your US ATLAS computing account name and password.


Powered by TWiki
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback