Re: Ceph-mon crashing FAILED assert(addr_name.count(p->second) == 0)

Looking at debug log, I don't see any unusual thing in accelio & xio messenger. I guess the problem happened because you installed MLNX_OFED_LINUX packages after compiling & installing accelio and ceph Installing MLNX_OFED package (with xen & kdm supports) will look for and install (re-install

XioMessenger howto

Hello Cephers, This is my cheat-sheet to build and bring up ceph/rdma. Please note that this cheat-sheet is for two nodes testing with ramdisk over RoCE. It installs Accelio lib and Ceph to non-standard locations (/opt/accelio and /opt/ceph). It uses *old mkcephfs* command to construct the clust

RE: xio messenger is not building by default

From: Somnath Roy [mailto:somnath@sandisk.com] Sent: Monday, March 16, 2015 11:59 AM To: Vu Pham; Matt W. Benjamin Cc: ceph-devel@vger.kernel.org; Sage Weil; Ken Dreyer Subject: RE: xio messenger is not building by default Vu, Yes, that means we can't use the apt-get version of lib

Re: xio messenger is not building by default

Hi Somnath, It's possible to build Accelio without OFED packages (download from open fabric website or MLNX_OFED is Mellanox version of the opensource package) installed; however, you need to install the libibverbs-dev, librdmacm-dev from the distribution you are running. If you compile agains

Re: [ceph-users] Ceph Hammer OSD Shard Tuning Test Results

I disabled ceph's header/data crc for both simplemessenger & xio but didn't run with header/data crc enable to see the differences. > >>>Were the simplemessenger tests on IPoIB or native? > >I think it's native, as the Vu Pham benchmark was done on mellanox &g

Re: xio messenger prelim benchmark

Hello Mark, >Hi Vu, > >Was that with auth enabled? I disabled auth on both run for both simple & xio (xio currently is not supporting cephx) This time I turned on max speed for all CPUs and get better results for both simple & xio > >On Feb 6, 2015 3:

Re: xio messenger prelim benchmark

On 2/6/2015 10:41:15 AM, "Vu Pham" wrote: > >>>>Yes, xio messenger which is implemented over Accelio can run over >>>>rdma >>>>transport (Infiniband, RoCE) and TCP. Please note that we have not >>>>enabled xio messenger / Acc

Re: xio messenger prelim benchmark

> Yes, xio messenger which is implemented over Accelio can run over rdma transport (Infiniband, RoCE) and TCP. Please note that we have not enabled xio messenger / Accelio-tcp yet. >> >>Oh, ok, Great ! >> >> xio messenger is currently working with user mode clients. We only >>>

Re: xio messenger prelim benchmark

>>>Yes, xio messenger which is implemented over Accelio can run over >>>rdma >>>transport (Infiniband, RoCE) and TCP. Please note that we have not >>>enabled xio messenger / Accelio-tcp yet. > >Oh, ok, Great ! > > >>>xio messenger is currently working with user mode clients. We only >>>validated/

Re[2]: xio messenger prelim benchmark

>Hi, > >I'm going to use same mellanox switchs in production next month, > >Isn't them ethernet switchs ? >How xio messenger is working, if they are not infiband ? > >Maybe with RoCE ? > Yes, xio messenger which is implemented over Accelio can run over rdma transport (Infiniband, RoCE) and TCP.

xio messenger prelim benchmark

Hi, I would like to share some benchmarking numbers on xio messenger and simple messenger HW/SW configuration: - . 1 32-core Xeon E5-2697V3 2.6G (Haswell) node, 64GB of memory. . Hyperthreading is ON/enabled, 64 cores . Mellanox ConnectX3-EN 40Gb/s HCAs, fw- 2.33.5000 . Mella

Re: Weekly performance meeting

Hi Sage, Thanks for this initiative. I'll definitely attend this weekly call. I'd like to discuss on how to make XioMessenger/rdma being useful and improving performance thanks -vu Hi everyone, A number of people have approached me about how to get more involved with the current work on i