LISTSERV mailing list manager LISTSERV 16.5

Help for XROOTD-L Archives


XROOTD-L Archives

XROOTD-L Archives


XROOTD-L@LISTSERV.SLAC.STANFORD.EDU


View:

Message:

[

First

|

Previous

|

Next

|

Last

]

By Topic:

[

First

|

Previous

|

Next

|

Last

]

By Author:

[

First

|

Previous

|

Next

|

Last

]

Font:

Proportional Font

LISTSERV Archives

LISTSERV Archives

XROOTD-L Home

XROOTD-L Home

XROOTD-L  May 2011

XROOTD-L May 2011

Subject:

Re: Resending: debugging poor performance of xroot across WAN

From:

Lukasz Janyst <[log in to unmask]>

Date:

16 May 2011 15:08:42 +0200Mon, 16 May 2011 15:08:42 +0200

Content-Type:

text/plain

Parts/Attachments:

Parts/Attachments

text/plain (151 lines)

Hi Charles,

2011/5/15 Charles G Waldman <[log in to unmask]>:
> #0  0xffffe430 in __kernel_vsyscall ()
> #1  0xf7712ef2 in pthread_cond_timedwait@@GLIBC_2.3.2 () from /lib/libpthread.so.0
> #2  0xef23d426 in XrdSysCondVar::Wait(int) ()   from .../libNetx.so
> #3  0xef19af48 in XrdClient::Read (this=0x16a44840, buf=0x1d70d600,offset=247982068, len=489) at XrdClient.cc:618
> #4  0xef2236d4 in TXNetFile::ReadBuffer(char*, int) ()  from .../libNetx.so
> #5  0xf586f623 in TBasket::ReadBasketBuffers(long long, int, TFile*)()from ../libTree.so
> #6  0xf58783fc in TBranch::GetBasket(int) () from .../libTree.so
>
>
> First off, note that the "XrdClient::Read" is only requisting 489 bytes in this read -
> no read-ahead or caching seems to be in effect.  As is well-known, many short reads
> like this will not perform well across the WAN.  So I am interested in testing the various
> read-ahead parameters, such as
>
>  XNet.ReadAheadSize
>  XNet.ReadAheadStrategy
>  XNet.ReadTrimBlockSize
>  XNet.ReadCacheSize
>
> Note that the system.rootrc for this job has:
>  XNet.ReadAheadSize: 0
>  XNet.ReadCacheSize: 10000000
>  XNet.ParStreamsPerPhyConn: 0

   There issue is a bit complicated here. The job that you are
analyzing is a ROOT job that uses XRootD plugin to access data over
the network. In your stack trace the entry point to XRootD is:
XrdClient::Read, everything above (#x < 3) is XRootD and everything
below (#x > 3) is ROOT. The parameters that you mention end up being
forwarded to the XRootD layer. Now, all the traffic optimization can
be done in two places: in XRootD or in ROOT:

1) When the traffic optimization is supposed to happen in XRootD, you
would see things like a XRootD::Read of len=489, because any external
application (ROOT in this case) just needs to state what data it needs
and the task of figuring out how to fetch it in the most optimal way
is entrusted to XRootD which will optimize the actual network reads.
The tweaking can then be done by the parameters you mention. Since
XRootD is a generic data access software it has no knowledge of the
underlying data file format so all it can do is some statistical guess
work that may be more or less optimal.

2) When the traffic optimization is supposed to happen in ROOT, you
would most probably see xrootd being asked to perform vector reads
(XrdClient::ReadV) of around 30 megs. This is far more optimal because
ROOT knows its file format and can easily predict which parts of the
file it will need in the nearest future so it is able to prefetch data
before it is needed by the application. The parameters here are the
ones that you mention below: tree->SetCacheSize and friends.

> Question 1)
> Why is ReadAheadSize set to 0 here?  And what's the best way to override this?

   Because most probably it is assumed that the optimization will be
done in ROOT and XRootD should not bother to do anything.

>     We'd love to turn it on by default. But the read-ahead needs to put the data somewhere: we need some extra memory; 30 to 100MB are enough. But with the experiments' never ending quest for memory resources turning this on by default is not an option: a 4-core machine with 4 jobs would eat an additional 400MB. Too much.

   Well, if you turn on the read-ahead or any prefetching you need to
store the additional data somewhere and RAM is the easiest target. We
work on some code that will hopefully be committed to ROOT soon which
will enable it to prefetch the data blocks and store them on disk.

>     Instead you need to turn it on by yourself by calling
>
>       tree->SetCacheSize(10000000);
>       tree->AddBranchToCache("*");
>
>
> I don't think we can force users to do this, is there somewhere else in the stack
> that this code could be inserted?

   Not really, since it's dependent on the user data that is being
read. You could argue that it should use some caching by default but
it's debatable.

> Question 2)
>
> Despite the small read-size and 0 read-ahead there may be something else happening as
> well - note from the attached strace output (which covers a time-span of 90 seconds)
> that there8373 messages sent to fd 79 (the xroot server) and no replies recieved.  All of the
> sent messages are 24 bytes in length:
>
>  send(79, "\3\0\v\305\0\0\0\0\0\0\0\0\16\306R\0\0\0\4\0\0\0\0\0", 24, 0) = 24
>
> these are going out at a rate of about 90 per second.   It seems that the messages
> are being sent but no replies recieved - in the logs on the xroot server I see:
>
>  110515 11:47:32 11957 XrootdXeq: usatlas1.12016:[log in to unmask] login
>  110515 11:47:32 11957 usatlas1.12016:[log in to unmask] XrootdProtocol: endsess 11955:61.163901
>  110515 11:47:32 11957 usatlas1.12016:[log in to unmask] XrootdProtocol: endsess 11955:61.163901 rc=-3 (No such process)
>  110515 11:47:32 11957 usatlas1.12016:[log in to unmask] XrootdResponse: sending err 3011: session not found
>
>
> Is this "session not found" the cause of the failures?
>
> Note that when this is happening, the server load is low (<1), the network is not
> congested, etc so there is no reason to be expecting timeouts.

   This is a request to end a session that apparently does not
exist... I will have a closer look. What is your access pattern? Do
you have long standing jobs that keep the connections open for a long
time, or is it more like the jobs fetching the data they need to
process and quiting after the processing is over?

> Question 3)
>
> I also see from strace output that the code is calling 'getrusage' excessively, does this
> really need to be checked 2500 times per second?
>
> Here's a count of system calls made in 90 seconds.
>  232452 getrusage
>  77484 gettimeofday
>  45394 time
>  16750 futex
>  15908 timer_settime
>   9131 clock_gettime
>   8373 send
>   8373 poll
>   1930 write
>   1137 lstat64
>    758 read
>    758 open
>    758 close
>     62 rt_sigaction
>     31 _llseek
>
> this seems quite inefficient!
>
> Note that the 758 instances of "read/open/close" are not recieving any data, they are just checking
> the processes memory usage:
>
>  open("/proc/self/statm", O_RDONLY|O_LARGEFILE) = 69
>  read(69, "223420 146018 5380 299 0 154771 "..., 46) = 34
>  close(69)                               = 0
>
> This is happening at a rate of ~10 Hz, which is not quite as bad as the ~2500Hz of "getrusage" calls.

   Are you sure that getrusage is called from XRootD and not from some
other place in your framework? I could not quite reproduce this issue
with xrdcp which uses the same underlying API.

Cheers,
   Lukasz



Top of Message | Previous Page | Permalink

Advanced Options


Options

Log In

Log In

Get Password

Get Password


Search Archives

Search Archives


Subscribe or Unsubscribe

Subscribe or Unsubscribe


Archives

March 2024
February 2024
January 2024
December 2023
November 2023
October 2023
September 2023
August 2023
July 2023
June 2023
May 2023
April 2023
March 2023
February 2023
January 2023
December 2022
October 2022
September 2022
August 2022
July 2022
June 2022
May 2022
April 2022
March 2022
February 2022
January 2022
December 2021
November 2021
October 2021
September 2021
August 2021
July 2021
June 2021
May 2021
April 2021
March 2021
February 2021
January 2021
December 2020
November 2020
October 2020
September 2020
August 2020
July 2020
June 2020
May 2020
April 2020
March 2020
February 2020
January 2020
December 2019
November 2019
October 2019
September 2019
August 2019
July 2019
June 2019
May 2019
April 2019
March 2019
February 2019
January 2019
December 2018
November 2018
October 2018
September 2018
August 2018
July 2018
June 2018
May 2018
April 2018
March 2018
February 2018
January 2018
December 2017
November 2017
October 2017
September 2017
August 2017
July 2017
June 2017
May 2017
April 2017
March 2017
February 2017
January 2017
December 2016
November 2016
October 2016
September 2016
August 2016
July 2016
May 2016
April 2016
March 2016
February 2016
January 2016
December 2015
November 2015
October 2015
September 2015
August 2015
July 2015
June 2015
May 2015
April 2015
March 2015
February 2015
January 2015
December 2014
November 2014
October 2014
September 2014
August 2014
July 2014
June 2014
May 2014
April 2014
March 2014
February 2014
January 2014
December 2013
November 2013
October 2013
September 2013
August 2013
July 2013
June 2013
May 2013
April 2013
March 2013
February 2013
January 2013
December 2012
November 2012
October 2012
September 2012
August 2012
July 2012
June 2012
May 2012
April 2012
March 2012
February 2012
January 2012
December 2011
November 2011
October 2011
September 2011
August 2011
July 2011
June 2011
May 2011
April 2011
March 2011
February 2011
January 2011
December 2010
October 2010
September 2010
August 2010
July 2010
June 2010
May 2010
April 2010
March 2010
February 2010
January 2010
December 2009
November 2009
October 2009
September 2009
July 2009
June 2009
May 2009
April 2009
March 2009
January 2009
December 2008
September 2008
August 2008
July 2008
June 2008
May 2008
April 2008
March 2008
February 2008
January 2008
December 2007
November 2007
October 2007
September 2007
August 2007
July 2007
June 2007
May 2007
April 2007
March 2007
February 2007
January 2007
December 2006
November 2006
October 2006
September 2006
August 2006
July 2006
June 2006
May 2006
April 2006
March 2006
February 2006
January 2006
December 2005
November 2005
October 2005
September 2005
August 2005
July 2005
June 2005
May 2005
April 2005
March 2005
February 2005
January 2005
December 2004
November 2004
October 2004
September 2004
August 2004

ATOM RSS1 RSS2



LISTSERV.SLAC.STANFORD.EDU

Secured by F-Secure Anti-Virus CataList Email List Search Powered by the LISTSERV Email List Manager

Privacy Notice, Security Notice and Terms of Use