Lustre 2.7.0 Released

We are pleased to announce that the Lustre 2.7.0 Release has been declared GA and packages are available for download.

A tarball of the source code is also available directly from the public git repository.

This major release includes the following new features:

  •  MDT-MDT Consistency Verification (LFSCK Phase 3) – Extends the FID-in-direct and linkEA verification from single MDT to cross-MDTs cases, and adds some inconsistent check/repair for remote object and striped directory. This development is funded by OpenSFS.
  • Dynamic LNET Config – Allows runtime configuration of Lustre networks, routes, and routers.
  • OST Placement – Provides fine-grained control of object placement via ‘lfs setstripe’. This development originated from Fujitsu.*
  • UID/GID Mapping -This is an IU developed feature that will allow mapping certain networks into pre-defined user/group ranges in order to avoid potential overlaps within different administrative domains. This is a technology preview that will be fully available in a future Lustre release.
  • DNE Striped Directory preview -While the striped directories are available in 2.7, it is not yet possible to do hard links or atomic rename() of files across MDTs within a striped directory, which will return -EXDEV. This will cause userspace tools like mv and rsync to make a copy of the file, but may be unexpected for some applications and users.

Fuller details can be found in the Changelog.

The following are known issues in the Lustre 2.7 Release:

  • LU-5939 – HSM agent operations might be replayed improperly in case of a server crash due to recovery. While producing error messages from HSM client at the time, this is safe since the agent will retry once the server is back up.
  • LU-6338 – Command lnetctl import del may show an error message even when the operation was successful.
  • LU-6340 – Sample lnet startup script may improperly configure Lustre routers configuration if both old and new style config information is present.

NOTE: Users should note that usage of the e2fsprogs-based lfsck has been deprecated and replaced by “lctl lfsck_start”. Using older e2fsprogs-based lfsck may lead to filesystem corruption. Once available, it is also recommended to use e2fsprogs-1.42.12.wc1 (or newer).

Please log any issues found through the issue tracking system.

We would like to thank OpenSFS for their contributions towards the cost of the release and also to all Lustre community members who have contributed to the release with code and/or testing.

LAD’14 – Lustre Administrators and Developers Workshop

September 22-23, 2014
Domaine Pommery, Reims – France

Organized by EOFS and OpenSFS in collaboration with the University of Reims Champagne-Ardenne

LAD’14 will take place in Reims, France, at Domaine Pommery during 2 days, on September 22-23, 2014. This will be a great opportunity for worldwide Lustre administrators and developers to gather and exchange their experiences, developments, tools, good practices and more!

Reims is reachable by train from Paris downtown and CDG Airport in about 45 minutes by the TGV, France’s high speed rail line.

We are waiting for your registration and presentation proposal!

Please use the following link to keep updated on LAD’14 agenda and logistics:

We are inviting community members to send proposals for presentations at this event. No proceeding is required, just an abstract of a 30-min (technical) presentation.
Please send this to [email protected] before July 25th, 2014.

Topics may include (but are not limited to): site updates or future projects, Lustre administration, monitoring and tools, Lustre feature overview, Lustre client performance (hot topic ;-), benefits of hardware evolution to Lustre (like SSD, many-cores…), comparison between Lustre and other parallel file systems (perf. and/or features), Lustre and Exascale I/O, etc.

Registration for the workshop is now open (early-bird rate):

We also recommend booking your hotel in downtown Reims as soon as possible.

On Monday evening, there will be a guided visit of Pommery Champagne cellars followed by a social dinner on site. A limited number of spouses can attend too (on a first-come, first-served basis). Register quickly!

This event is organized thanks to the following generous sponsors:
CEA, DataDirect Networks and Intel

We expect more sponsors to be confirmed soon. Please contact [email protected] if your company wants to sponsor LAD.

For any other information, please contact [email protected]

New Report from the Benchmarking Working Group

I/O Characterization of Large-Scale HPC Centers — Benchmarking Working Group, OpenSFS (PDF)

The  OpenSFS  Benchmarking  Working  Group  (BWG)  was  created  with  the  intent  of  defining  an  I/O  benchmark suite to satisfy the requirements of the scalable parallel file system users and facilities. The  first step toward this end was identified as characterization of I/O workloads, from small scale to very large scale parallel file systems, deployed at various highCperformance and parallel computing (HPC) facilities  and  institutions.  The  characterization  will  then  drive  the  design  of  the  I/O  benchmarks  that  emulate  these workloads.

As part of the characterization, the BWG released a survey at the Supercomputing Conference in 2012,  to  solicit  participation  and  collect  data  on  file systems  and  workloads  in  HPC  centers.  This  paper  summarizes the data collected and our analysis.

Lustre Strategy Survey Open

The OpenSFS Community Development Working Group is gathering data from organizations using Lustre in order to develop a long-term support strategy recommendation going forward. We want to ensure that future Lustre releases are well-aligned with the needs of the Lustre community.

Please complete this short survey to make sure that your organization’s voice is heard!

Note that all questions are optional you may submit a partially completed survey if you prefer.