[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-users] Using Xen Virtualization Environment for Development and Testing of Supercomputing and High Performance Computing (HPC) Cluster MPICH2 MPI-2 Applications

  • To: xen-devel@xxxxxxxxxxxxxxxxxxx, xen-users@xxxxxxxxxxxxxxxxxxx
  • From: "Mr. Teo En Ming (Zhang Enming)" <space.time.universe@xxxxxxxxx>
  • Date: Fri, 30 Oct 2009 14:37:17 +0800
  • Cc: space.time.universe@xxxxxxxxx
  • Delivery-date: Thu, 29 Oct 2009 23:38:21 -0700
  • Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:cc:content-type; b=sZErgoS1hDqTBCvvGXvJ9GmVOdtUR+s4+HPvR3xFq0c9lEKHjIlJBwSJT3Tb1hTXB7 71NXbfXyL7t0Y8AwtuC/2tOafW6Uw7rgmU9n6Mm4NiG+Vx5hjFKVzY1YJOfs5Fj2fmd8 0D5mzHizAKT8ilWqPQVIcJ/Ys+xAvdyzYj6GI=
  • List-id: Xen user discussion <xen-users.lists.xensource.com>

Dear All,

I have created a virtual high performance computing (HPC) cluster of 6 compute nodes with MPICH2 using Xen-based Fedora 11 Linux 64-bit paravirtualized (PV) domU guests. Dom0 is Fedora 11 Linux 64-bit. My Intel Desktop Board DQ45CB has a single onboard Gigabit LAN network adapter.

I am able to bring up the ring of mpd on the set of 6 compute nodes. However, I am consistently encountering the "(mpiexec 392): no msg recvd from mpd when expecting ack of request" error.

After much troubleshooting, I have found that there are Receive Errors (RX-ERR) in the virtual network interface eth0 of all the six compute nodes. All the 6 compute nodes are identical F11 linux 64-bit PV virtual machines.

Here is my PV guest configuration for node 1:

[enming@fedora11-x86-64-host xen]$ cat enming-f11-pv-hpc-node0001
disk = ['phy:/dev/virtualmachines/f11-pv-hpc-node0001,xvda,w' ]
vif = [ 'mac=00:16:3E:69:E9:11,bridge=eth0' ]
vfb = [ 'vnc=1,vncunused=1,vncdisplay=0,vnclisten=,vncpasswd=' ]
bootloader = "/usr/bin/pygrub"
#kernel = "/home/enming/fedora11/vmlinuz"
#ramdisk = "/home/enming/fedora11/initrd.img"
Will there be any problems with Xen networking for MPICH2 applications? Or it's just a fine-tuning exercise for Xen networking? I am using PV guests because PV guests have much higher performance than HVM guests.

Here are my mpich-discuss mailing list threads:






Please advise on the RX-ERR.

Thank you very much.

Mr. Teo En Ming (Zhang Enming) Dip(Mechatronics) BEng(Hons)(Mechanical Engineering)
Alma Maters:
(1) Singapore Polytechnic
(2) National University of Singapore
My blog URL: http://teo-en-ming-aka-zhang-enming.blogspot.com
My Youtube videos: http://www.youtube.com/user/enmingteo
Email: space.time.universe@xxxxxxxxx
MSN: teoenming@xxxxxxxxxxx
Mobile Phone (SingTel): +65-9648-9798
Mobile Phone (Starhub Prepaid): +65-8369-2618
Age: 31 (as at 30 Oct 2009)
Height: 1.78 meters
Race: Chinese
Dialect: Hokkien
Street: Bedok Reservoir Road
Country: Singapore
Xen-users mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.