MD BMG: Difference between revisions

From OpenSFS Wiki
Jump to navigation Jump to search
No edit summary
No edit summary
 
(30 intermediate revisions by 3 users not shown)
Line 1: Line 1:
The task of the BWG Metadata Performance Evaluation Effort (MPEE) group is to:
[[Benchmarking_Working_Group|Back to the main BWG page]]
# Build/select tools that will allow evaluation of File System Metadata performance and scalability
# The tools will help detect pockets of Metadata low performance in cases when users complain of extreme slowness  of MD operations
# Benchmark tools will support: POSIX, MPI, and Transactional operations (for CEPH and DAOS)
# Address the very high end HPC as well as small and medium installations benchmark needs
# Tools applicable to Lustre and: CEPH, GPFS…


'''Lead:''' Sorin Fabish <br />
'''Members:''' Branislav Radovanovic, Rick Roloff, Cheng Shao, Wang Yibin, Keith Mannthey, Bobbie Lind, Greg Farnum
<br />


'''Current MPEE proposed list of benchmarks''':  
[[MD_BMG/charter|'''The task of the BWG Metadata Performance Evaluation Effort (MPEE)''']] group is to:
# mdtest – widely used in HPC
<br \> Build/select tools that will allow evaluation of File System Metadata performance and scalability
# fstest - used by pvfs/OrangeFS community
# Postmark and MPI version - old NetApp benchmark
# Netmist and MPI version – used by SPECsfs
# Synthetic tools – used by LANL, ORNL
# MDS-Survey - Intel’s metadata workload simulator.
# Any known open source metadata tools used in HPC
# Add new Lustre statistics specific to MD operations.


'''MPEE Usecases'''
[[MD_BMG/benchmarks|'''Current MPEE proposed list of benchmarks''']]
* '''mdtest''': test file MD operations on MDS: open, create, lookups, readdir; used in academia and as a comparison tool of FS MD.
* '''fstest''': small I/O’s and small files as well as lookups, targeting both MDS and OSS operations and MD HA for multiple MDS’s.
* '''Postmark''': old NetApp benchmark – I built an MPI version; it is used to measure MD operations and file size scalability and files per directory scalability.
* '''Netmist''': used to model any workload from statistics including all MD operations and file operations. Can model Workload objects for I/O performance mixes and combination of I/O and MD. Suitable for initial evaluation of storage as well as for performance troubleshooting.


'''MPEE Proposed Roadmap'''
[[MD_BMG/Usecases|'''MPEE Usecases''']]
* Collect benchmark tools candidates from OpenSFS
* Evaluate all the tools and the workloads that can benchmarked
* Recommend a small set of MD benchmark tools to cover the majority of MD workloads
* Collect stats from users of MD benchmarks 
* Build scripts to allow ease of use of the recommended tools
* Write documentation for troubleshooting MD performance problems using the toolset 
* Create a special website for MD tools


'''MPEE Asks from OpenSFS'''
[[MD_BMG/roadmap|'''MPEE Proposed Roadmap''']]
* Share any open source synthetic benchmarks code
<br />
* Share a list of MD benchmark tools they currently use to allow select the most suitable and used candidates
At SC'13 we will present results of running the 4 benchmark candidates on same system, compare their results,
* Share MD operations tested to allow build Netmist workload objects
<br />and discuss their eligibility as proposed benchmarks of MPEE.
* Share the MD workloads that create pain points to Lustre FS
<br />
* Share cases of poor MD performance workloads and applications
* [http://wiki.opensfs.org/images/4/46/LUG14_MPEE_BWG_OpenSFS.pdf '''Metadata Benchmarks and MD Performance Metrics'''] This presentation was approved to be presented at LUG 2014. Any review is appreciated.([[MD_BMG/LUG-abstract|'''See Abstract''']])
<br />
[[MD_BMG/asks|'''MPEE Asks from OpenSFS''']]:
<br />
Share any open source synthetic benchmarks code


Return to [[http://wiki.opensfs.org/Benchmarking_Working_Group Benchmarking_Working_Group]] page.
Return to [[http://wiki.opensfs.org/Benchmarking_Working_Group Benchmarking_Working_Group]] page.

Latest revision as of 11:56, 6 March 2015

Back to the main BWG page

Lead: Sorin Fabish
Members: Branislav Radovanovic, Rick Roloff, Cheng Shao, Wang Yibin, Keith Mannthey, Bobbie Lind, Greg Farnum

The task of the BWG Metadata Performance Evaluation Effort (MPEE) group is to:
Build/select tools that will allow evaluation of File System Metadata performance and scalability

Current MPEE proposed list of benchmarks

MPEE Usecases

MPEE Proposed Roadmap
At SC'13 we will present results of running the 4 benchmark candidates on same system, compare their results,
and discuss their eligibility as proposed benchmarks of MPEE.


MPEE Asks from OpenSFS:
Share any open source synthetic benchmarks code

Return to [Benchmarking_Working_Group] page.