In the meeting this morning I said I thought that the
roadrunner system has the largest IB fabric with 6-7K nodes,
but that was incorrect.

It was actually the NASA system that I was thinking of, that has 6400 nodes.

http://www.nas.nasa.gov/Resources/Systems/pleiades.html

________________________________
From: [email protected] 
[mailto:[email protected]] On Behalf Of Tziporet Koren
Sent: Monday, November 09, 2009 8:26 AM
To: OpenFabrics EWG; Tziporet Koren
Subject: [ewg] EWG/OFED agenda for today (Nov 09, 2009)


OFED Meeting agenda for today:

1. OFED 1.5 release status:
    - RC2 was released last week
    - Bugs status review: We have many blocker and critical bugs - how can we 
progress faster?

2. Open discussion (maybe Woody wish to review the OFA slides for SC09)

Bugs list:

1671    blocker Other   
[email protected]<mailto:[email protected]>     NFSRDMA 
backport headers being used even when NFSRDMA is not being used
1743    blocker CentOS 5        
[email protected]<mailto:[email protected]> 
NFSoverRDMA-iozone test stalls on directory mounted with NFS over RDMA on 
Centos5.3 64 bit
1764    blocker CentOS 5        [email protected]<mailto:[email protected]> 
error: unknown field ‘eh_target_reset_handler’ specified in initializer
1778    blocker CentOS 5        
[email protected]<mailto:[email protected]> OFED-1.5-IMB 
tests Alltoall & Alltoallv aborts while running with -npmin32 with OpenMPI on 8 
nodes
1819    blocker CentOS 5        
[email protected]<mailto:[email protected]> OFED-1.5-MTT 
trivial tests fails to complete on Centos5.3-x86_64 bit platform
1800    critical        Other   
[email protected]<mailto:[email protected]>       iperf sdp on ppc cause 
to client machine to dead lock
1805    critical        RHEL 5  
[email protected]<mailto:[email protected]>     Kernel panics 
when NFS server is 32 bit OS & NFS client is 64 bit OS while trying to mount 
shared NFS directory.
1810    critical        RHEL 5  
[email protected]<mailto:[email protected]>     OFED-1.5- 
NFSoverRDMA- "Lock tests" of Connectathon fails to run on RHEL5U2-32 bit 
platform
1811    critical        RHEL 5  
[email protected]<mailto:[email protected]>     OFED-1.5- 
NFSoverRDMA-"SPECIAL TESTS" of Connectathon tool fails on 2.6.30 -32 bit kernel
1812    critical        SLES 11 
[email protected]<mailto:[email protected]>       SDP fails during init_qp
1824    critical        Other   
[email protected]<mailto:[email protected]>     Call trace on 
NFSoRDMA server
1636    major   SLES 11 [email protected]<mailto:[email protected]> SLES10 
and SLES11 expect a -32bit suffix for any user space 32bit RPMs on 64 bit CPU 
platforms.
1719    major   Other   
[email protected]<mailto:[email protected]>     Reconnect 
path in Linux client can loop indefinitely
1789    major   CentOS 5        
[email protected]<mailto:[email protected]>     OFED-1.5 
kernel panic observed while running iozone & connectathon concurrently.
1793    major   Other   [email protected]<mailto:[email protected]>       
Netperf sdp with msg size 128 or 256 cause to call trace in dmesg and the 
machine to stuck
1815    major   RHEL 5  
[email protected]<mailto:[email protected]>     NFSoRDMA: 
mount.rnfs on client side does not work
1821    major   RHEL 5  [email protected]<mailto:[email protected]>   Crash 
in bonding
1823    major   SLES 11 [email protected]<mailto:[email protected]> 
OFED-1.5-rc2 fails to discover SRP direct connect targets

_______________________________________________
ewg mailing list
[email protected]
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg

Reply via email to