From admin at opensfs.org Fri Aug 21 16:01:29 2015 From: admin at opensfs.org (OpenSFS Administration) Date: Fri, 21 Aug 2015 09:01:29 -0700 Subject: [Openbenchmark] August 2015 OpenSFS Newsletter Message-ID: <000501d0dc2a$a4056ae0$ec1040a0$@opensfs.org> OpenSFS is excited to introduce our new newsletter. We hope you enjoy the following articles: - ISC Showcases Lustre Worldwide - OpenSFS News - Lustre 2.8 - Coming Soon ISC Showcases Lustre Worldwide: The International Supercomputer Conference 2015 ,recently held in Frankfurt, Germany, is the largest annual supercomputing show outside of North America. Every year it does a great job showcasing advances by European companies and products in the field of supercomputing. This year, the topics ranged widely under the stated rubric of diversity, including the Top3 Exascale Projects, HPC in the Global Automotive Industry, Future HPC System Design Concepts, HPC and Cloud Computing in Financial Services, and even the Potential Impact of Quantum Computing On HPC. OpenSFS participants were in full force with a spate of Lustre-focused announcements. For the full newsletter content, including information on Lustre 2.8, how to get involved with Lustre for free, and more, please visit our web site and review the online version of the newsletter . We encourage you to contact OpenSFS Administration with future newsletter content recommendations. __________________________ OpenSFS Administration 3855 SW 153rd Drive Beaverton, OR 97003 USA Phone: +1 503-619-0561 | Fax: +1 503-644-6708 Twitter: @OpenSFS Email: admin at opensfs.org | Website: www.opensfs.org -------------- next part -------------- An HTML attachment was scrubbed... URL: From tiwari at ornl.gov Tue Aug 25 21:05:19 2015 From: tiwari at ornl.gov (Tiwari, Devesh) Date: Tue, 25 Aug 2015 21:05:19 +0000 Subject: [Openbenchmark] Feedback for OpenSFS Board Meeting Message-ID: All, As we had discussed, we plan to bring up some new items in tomorrow's board meeting. I could pick Andrew's brain on these issues as well. He has supported the idea and made some great suggestions. Based our previous meetings and my interaction with Andrew, this is what I have come up with. Please provide feedback, esp. for the ideas that we could implement (we currently have five items), but better suggestions can augment/replace the current items. I am thinking of limiting it to three items though. -------- Report status update: o Regular BWG meets bi-weekly. o All meetings took place. Nothing canceled. o Submitted BoF to the SC. The WG discussed extensively about what should go out in the BoF, what we should discuss etc. o We reached out to experts in the area and got some positive response. Followup (leading to the real discussion/proposal point): - The attendance in the bi-weekly meetings has been somewhat mild. Perhaps due to combination of multiple reasons, including not-my-day-job, summer/vacation time, job changes, focus on one task only (hero-run), etc. BWG has started an introspection phase where we are looking at how we could get more visibility and impact the community in more meaningful ways. We are looking at ways to improve the engagement and potential impact. Proposal: o We should rebrand ourselves as "Performance Working Group". Likely to attract more participations. A platform where both user and vendor community can engage in a meaningful conversation. Participants can learn and share tricks, tools and deep technical knowledge. Likely to result in more tangible and meaningful outcomes. Some ideas that we could implement. 1. Top500 list for storage systems. Similar to Top500 for supercomputers. 2. File and Storage system bibliography (with short summary and keywords that describe the paper) -- could be very useful to the community at-large. 3. File and Storage system on-line journal (researchers submit their articles and we publish after reviewing, similar to arxiv). 4. Interesting technology evaluation/products related collection. We could pull interesting stuff about storage from Slashdot/conferences/AnandTech etc. and put them at one place. 5. Community could share some tools/log/trace/performance data on our wiki. I think there is already some website where researchers collect traces etc. But, we could collect more useful information like failure logs, MDS request information etc. o However, implementing any of these ideas requires some serious effort. So, we will need think of how to gather dedicated resources for these efforts as we move along. …∂t -------------- next part -------------- An HTML attachment was scrubbed... URL: