I forgot to include the operation mix for the workload objects included in the results.
FS Operation SFS2008 BACKUP HOMEDIR DATABASE STREAM VIRTUAL read 10 9 8 1 10 7 mmap read 0 0 0 0 0 0 rand read 8 9 10 79 65 20 write 7 7 7 2 10 13 mmap write 0 0 0 0 0 0 rand write 2 2 2 18 2 60 rmw 1 1 1 0 1 0 mkdir 2 2 2 0 1 0 unlink 4 4 4 0 1 0 append 3 3 3 0 1 0 locking 6 6 6 0 1 0 access 24 24 18 0 1 0 stat 26 26 20 0 5 0 chmod 2 2 3 0 1 0 readdir 4 4 5 0 1 0 copyfile 0 0 10 0 0 0 rename 0 0 0 0 0 0 statfs 1 1 1 0 0 0 pathconf 0 0 0 0 0 0 custom1 0 0 0 0 0 0 custom2 0 0 0 0 0 0 _____________________________________________ From: faibish, sorin Sent: Friday, May 18, 2012 12:06 PM To: [email protected]; [email protected]; [email protected]; 'Oral, H. Sarp' Subject: RE: Benchmarking Working Group Oral, As I promised I ran all the workload types with the netmist (next sfs2013 benchmark) benchmark with the existent workload objects. I am currently working on defining a new Lustre workload object but I sent this to the group as a reference. Attached please find a document including all the workload types results for a small system (details included in the document). I included the details on the workload object at the end of the pdf: << File: netmist_lustre_60drives.pdf >> Thank you for your understanding and support Sorin Faibish Distinguished Engineer Fast Data Group - CTO Office EMC -----Original Appointment----- From: Vanderbilt, Richard [mailto:[email protected]] Sent: Monday, March 19, 2012 12:25 PM To: [email protected]; [email protected]; [email protected]; faibish, sorin Subject: [Openbenchmark] Benchmarking Working Group When: Friday, May 18, 2012 11:30 AM-12:30 PM (UTC-05:00) Eastern Time (US & Canada). Where: 1(866)844-9418 id 4088226016 The OpenSFS Benchmarking Workgroup aims to provide an I/O benchmark suite to satisfy the benchmarking requirements of the scalable parallel file system users and facilities. Towards this end, as a first step, the workgroup aims to characterize the I/O workloads from small- to very large-scale scalable parallel file systems deployed at various high-performance computing and parallel computing facilities and institutions. Using these gathered characteristics, the workgroup will then collect and build the required I/O benchmarks to emulate these workloads and provide required documentation about the benchmark suite. The workgroup meeting will reoccur every Friday at 8:30am Pacific time. 1 hour each. The dial-in is as follows: 1(866)844-9418 id# 4088226016. Also note: we will be having our first face to face meeting on Sunday April 22nd from 2:00-5:00pm at the LUG. Details to come. On Tuesday April 24 4:10-5:00pm there will be a Benchmark panel discussion moderated by Sarp Oral. Panelist will be Jeremy Filizetti (DOD), Andrew Uselton (NERSC), Eric Barton (Whamcloud), Nathan Rutman (Xyratex), Stephane Thiell (CEA), and Stephen Simms (IU). If you are interested in contributing to this work group, please email Sarp Oral at [email protected]<mailto:[email protected]> or myself at [email protected]<mailto:[email protected]>. Sincerely, Sarp Oral, PhD National Center for Computational Sciences Oak Ridge National Laboratory [email protected]<mailto:[email protected]> 865-574-2173 Richard Vanderbilt Co-chair Netapp Inc. (408)822-6016 << File: ATT00001.txt >>
_______________________________________________ Lustre-discuss mailing list [email protected] http://lists.lustre.org/mailman/listinfo/lustre-discuss
