From ggrider at lanl.gov Wed Jan 17 12:25:50 2007 From: ggrider at lanl.gov (Gary Grider) Date: Wed, 17 Jan 2007 10:25:50 -0700 Subject: [Storage-research-list] Fwd: DOD NSA Advanced Computing Systems Visiting Research Scholar Sabbatical Program Message-ID: <6.2.3.4.2.20070117102510.050301a0@cic-mail.lanl.gov> > >Most of you probably know about the DOD NSA ACS Program, but for >those that dont, >the program will be looking for R&D projects within >industry/academia/govt labs to fund >that will move technology closer to usable and to break paradigms >that wouldnt be broken otherwise. >This will be done mostly through Government Broad Area Announcements. > >Additionally, the program may have a few visiting paid research >scholars (on Sabbatical from >a normal assignment). The visiting scholar would be paid to think >out of the box on challenging >problems of interest to the NSA. These positions would be temporary >positions allowing top >people to step away from current tasks in the tradition of a >Sabbatical. DOD clearances and US >citizenship are NOT required. > >One of the thrust areas of the ACS Program is File Systems, I/O, and >Storage in >high performance and high bandwidth computing applications. > >If you are looking for an opportunity to be a visiting scholar and >be given a paid chance >to think about the some of the most challenging problems in our >area, please contact me. > >Thanks >Gary Grider >Los Alamos National Laboratory/DOE NNSA >DOD NSA ACS Program FSIO thrust coordinator >ggrider at lanl.gov From renbyna at mcs.anl.gov Sun Jan 28 10:38:39 2007 From: renbyna at mcs.anl.gov (Suren Byna) Date: Sun, 28 Jan 2007 09:38:39 -0600 Subject: [Storage-research-list] looking for applications with many I/O read requests Message-ID: <991297875.20070128093839@mcs.anl.gov> Hi, I'm Surendra Byna from Illinois Institute of Technology, Chicago. We are working on a project for improving performance of I/O intensive applications. We are looking for sample applications, benchmarks, and I/O tracefiles that have a large number of read requests. Could you please let us know if you have any such applications or information on where I can look at. Thank you. -- Best regards, Suren mailto:renbyna at mcs.anl.gov From steven.w.schlosser at intel.com Mon Jan 29 12:12:41 2007 From: steven.w.schlosser at intel.com (Schlosser, Steven W) Date: Mon, 29 Jan 2007 10:12:41 -0700 Subject: [Storage-research-list] Intel Research Pittsburgh is looking for summer interns! Message-ID: <25F787474E5ACF4BB7E3ABF6C4A396F801596B9B@azsmsx413.amr.corp.intel.com> Intel Research Pittsburgh invites applications for internship positions from experimental computer systems researchers. We are looking for students who combine excellent hands-on systems skills with the ability to conceptualize innovatively in core systems areas. Our research areas include Computer Architecture Operating Systems Compilers & Binary Translation Distributed Systems Computer Systems Simulation Networking Virtualization File Systems Parallel Processing Storage Systems Robotic Manipulation Databases Modular Robotics Run-time Verification Multi-Agent Motion planning Medical Image Processing Computer Vision Algorithms Machine Learning We are specifically looking for interns in several of the lab's research efforts, including Dynamic Physical Rendering (DPR), Log-Based Architectures (LBA), Reliable Email (Re:); Interactive Search-Assisted Diagnosis (ISAD); novel parallel programming models; Personal Robotics, Data-Oriented Transfer (DOT); and personal data management. Other project opportunities are also available. For more details, see http://www.pittsburgh.intel-research.net/ . As part of Intel's network of university-affiliated labs, we conduct research in an open, collaborative manner and encourage prompt publication of research results. The software developed in our research efforts is typically disseminated in open-source form. Intel Research Pittsburgh is located in the Collaborative Innovation Center on the Carnegie Mellon University campus. Lab researchers typically collaborate closely with the university's Computer Science and ECE departments. The lab is located in the Pittsburgh's vibrant Oakland district, home to the University of Pittsburgh, the Carnegie Museum, and several parks. Numerous outdoors activities are available throughout the area. To apply for an internship position, please visit: http://www.pittsburgh.intel-research.net/internship2007/ The application deadline is February 16, 2007 (applications are considered as they arrive). -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdigney at cs.cmu.edu Tue Mar 20 13:47:05 2007 From: jdigney at cs.cmu.edu (Joan Digney) Date: Tue, 20 Mar 2007 11:47:05 -0600 Subject: [Storage-research-list] Research Intern Positions at Seagate Pittsburgh Message-ID: <46001E19.5090504@cs.cmu.edu> Sent on behalf of Erik Riedel, Seagate, Pittsburgh ----- research intern positions @ Seagate (Pittsburgh, PA) The Interfaces and Architecture department at Seagate Research is looking for multiple research interns for the Spring and Summer of 2007. Requirements are for highly motivated and independent students who are currently pursuing their Ph.D. or M.S. work in the area of computer systems or a related field (2nd year or beyond preferred). Direct research experience with storage systems or file systems is a plus, but a background in related areas such as operating systems, networking, performance evaluation, mobile computing or databases is also welcome. Seamless Movement of Digital Media - understand the underlying technical issues in the communication among consumer devices to exchange and synchronize stored data. The goal of this project is to better understand how today's media-rich devices are talking to each other - e.g., digital cameras, camcorders, TVs, PVRs, PDAs and laptops - and to identify the technical barriers (not the marketplace artifacts) to seamless interoperability among all the gadgets that play a central role in managing people's data. Why does it seem so difficult to do? Can existing technology from disconnected operation, distributed systems or peer-to-peer neatly address all the issues? Skills required include a knowledge of networking, file systems, peer-to-peer and multimedia. Data Mining for Workload Characteristics - identify data mining techniques to organize, analyze and characterize multivariate time series. An example is a data set of disk request arrivals over a long period of time from a large number of drives. Questions to be answered include: What type of data clustering techniques are effective in such cases? Which metrics should be used to compactly characterize such a large data set? What size populations are required to draw strong conclusions? Monitoring and Self-Management - In self-managed systems, it is common to deploy feedback loops for system management such that the system parameters are tuned to current conditions. This requires monitoring of a system's performance. Another approach is to monitor the system conditions themselves, characterize them compactly (e.g. using histograms or similar structures) and tune algorithm parameters based on those observations. This allows the system to adapt to changes in a timely fashion with emphasis on efficiency and robustness. The goal of this project is to compare and evaluate multiple possible algorithms for use in an adaptive storage system. CONTACT (please reference one of the job areas above): Lori Beal, Seagate Recruiting Lori.A.Beal at seagate.com Erik Riedel, Seagate Research Erik.Riedel at seagate.com http://www.seagate.com/jobs ABOUT SEAGATE RESEARCH: Seagate Research, located in Pittsburgh, Pennsylvania is the central research lab for all of Seagate worldwide. The groups at Research are responsible for the future technology advances that will be needed in Seagate products 4 to 10 years in the future. Our work covers the entire spectrum of what it takes to succeed at disc drive engineering, the "extreme sport" of the technology industry. Our staff includes experts in magnetic recording, mechanical engineers, material scientists, physicists, chemists, computer engineers, computer scientists, mathematicians and modelers all working on technology to cover the entire spectrum of Seagate products from portable consumer devices to mission-critical enterprise storage. ABOUT SEAGATE TECHNOLOGY: Seagate was named Forbes magazine "Company of the Year" for 2005, chosen from among 1000 candidate companies judged on multiple criteria, including total return, sales and earnings growth, innovation and market leadership. See the January 9th, 2006 issue. http://www.forbes.com/free_forbes/2006/0109/092.html Seagate is the world?s leading provider of storage technology for Internet, business and consumer applications. The Company?s products include disc drives for the Enterprise, PCs and Consumer Electronics, as well as Storage Area Network (SAN) solutions and Server Appliances. Seagate?s market leadership is based on delivering award-winning products, customer support and reliability to meet the world?s growing demand for information storage. Seagate can be found around the globe and at www.seagate.com ABOUT PITTSBURGH: Pittsburgh has been called the "Epicenter of Storage Innovation" with a long record as a leader in storage innovations that have spread out from here and left their mark on the entire industry. This stretches from AFS and Coda research at Carnegie Mellon and later Transarc and IBM to storage research at the Pittsburgh Supercomputer Center (PSC), to numerous innovations from the Data Storage Systems Center (DSSC) and the Parallel Data Laboratory (PDL). Intel and Seagate both have storage-focused research labs in Pittsburgh. Network Appliance and startup Panasas also have design centers in Pittsburgh. From elm at cs.ucsc.edu Tue Mar 27 21:13:25 2007 From: elm at cs.ucsc.edu (Ethan L. Miller) Date: Tue, 27 Mar 2007 18:13:25 -0700 Subject: [Storage-research-list] CFP: 2007 IEEE Conference on Mass Storage Systems & Technologies (MSST 2007) Message-ID: Paper submission deadline is April 30, 2007. -------- IEEE MSST2007 Twenty-Fourth IEEE Conference on Mass Storage Systems and Technologies MSST2007 will be a venue for presenting the latest research and development in many aspects of large-scale and long term storage. The conference will focus on three specific challenges for mass storage systems: global-scale distributed storage, long-term data storage and stewardship, and large-scale high-performance storage systems. Specific areas of interest for MSST2007 include developments in storage technologies such as disk storage, tape storage, clustered storage, and other emerging technologies; advances in high speed global networks and protocols to support storage; security and metadata management in global-scale, high-performance, and archival storage; techniques for building scalable storage systems; approaches to long-term data preservation and management; and performance modeling and analysis of large storage systems. "Idea" papers that discuss potential approaches, proof-of-concept papers that show the effectiveness of potentially limited-scale implementations, papers that describe real-world experience with large-scale storage systems are all welcome. How to Participate We encourage participation by individuals and organizations with research or practical experience, including government and commercial users, universities, research laboratories, and early adopters. We invite submissions of technical papers, tutorial proposals, short paper/poster presentations, and vendor briefings. Technical Papers and Poster Presentations: Submit a full 8-14 page paper or short 3-5 page paper online at http://storageconference.org/ by April 30, 2007. Papers should be in two columns with 10 point font and should be in PDF; specific formatting details are available at the submission site. All papers will receive full, detailed reviews from at least 3 PC members (in contrast to previous conferences), with acceptance notification provided by June 22, 2007. Final camera-ready papers, incorporating reviewer comments, must be submitted in PDF format no later than July 20, 2007, and must conform to the formatting standards described in the Author's Kit, which will be available at the conference site. Full 8-14 page papers will be considered for both full 30 minute presentations and the poster session; please indicate on your submission form if you do not want your paper to be considered for the poster session. Short paper submissions of 3-5 pages in length will only be considered for the poster session. Questions about technical papers or possible submissions may be directed to the program chair, Ethan Miller, at pcchair07 at storageconference.org. Tutorial Proposals: There will be several tutorial sessions at MSST held on September 24, 2007. Submit outlines of half- or full-day tutorials, with material to be covered and a biography of the proposed tutorial presenter by April 30, 2007 via email to tutorial07 at storageconference.org. Vendor Briefings: We will provide a block of time during the conference for vendors to present information addressing the conference theme. Submit a 3-5 page abstract by April 30, 2007 via email to vendor07 at storageconference.org. Vendor Expo: Space for vendor product exhibits will be assigned in the order reservations are received. Space is limited. Vendor inter-operability demonstrations are encouraged. To register and to discuss facilities and cost, contact the Vendor Expo Chair, Ben Kobler, at ben.kobler at nasa.gov. Questions that aren't covered by specific people listed above may be directed to James Hughes, the general chair, at generalchair07 at storageconference.org. General chair: James Hughes, Sun Microsystems Program chair: Ethan L. Miller, University of California, Santa Cruz Vendor Expo chair: Ben Kobler, NASA Goddard Space Flight Center Program committee: Ahmed Amer, University of Pittsburgh Curtis Anderson, Data Domain Jean-Jacques Bedet, Adnet Systems Greg Bronevetsky, Lawrence Livermore National Laboratory Randal Burns, Johns Hopkins University Bob Chadduck, NARA John Chandy, University of Connecticut Jack Cole, US Army Research Laboratory Zoran Dimitrijevic, Google Gene Harano, NCAR James Hughes, Sun Microsystems Merritt Jones, MITRE Ben Kobler, NASA Goddard Space Flight Center Reagan Moore, San Diego Supercomputer Center James Nunez, Los Alamos National Laboratory Matthew O'Keefe, Cray Ron Oldfield, Sandia National Laboratory Tom Ruwart, University of Minnesota Julian Satran, IBM Haifa Research Laboratory, Israel Thomas Schwarz, Santa Clara University Rodney Van Meter, Keio University, Japan Theodore Wong, IBM Research Jay Wylie, Hewlett Packard Qin Xin, Symantec From coyne at us.ibm.com Mon Apr 16 17:09:49 2007 From: coyne at us.ibm.com (Bob Coyne Jr) Date: Mon, 16 Apr 2007 16:09:49 -0500 Subject: [Storage-research-list] Call For Participation IEEE Archive Life Cycle Workshop Message-ID: Folk, please consider participating. Thanks,Bob Coyne IEEE MSSTC September 24, San Diego, California, USA Archival Storage Life Cycle Management A Workshop at the Twenty-Fourth IEEE Conference on Mass Storage Systems and Technologies Workshop Description The one-day workshop will examine the issues and approaches to providing data stewardship in very large data environments where copying all the data from one generation system to another is not feasible. There are multiple technical, policy, legal and other related life cycle matters that affect data stewardship. Examples are: media aging and reliability characteristics; evolution over time of file system and archive software; standards; government initiatives, programs contracting trends; laws that affect retention management; security policies and technology; and the general evolution of storage technology. These matters will be discussed and expanded in terms of importance and difficulty, and the solution space will be qualified where time and agreement allow. The effort is intended to be a vendor-neutral exercise. The workshop will take place on Monday, September 24, 2007, the day before the beginning of general sessions of the IEEE Mass Storage Systems and Technology Conference MSST2007. See the conference web site: www.storageconference.org/2007 A short white paper is requested from each participant in the workshop, as described below. The white papers will be correlated before the workshop to create a framework for considering approaches and possible solutions. The workshop will generate a report to be published by the IEEE Computer Society Press with a list of participants included, and a summary of the report will be presented during the conference. How to Participate Send email requesting participation to: archivestorage at ieee.org Participants are asked to submit white papers by June 24, 2007. The white paper will be based on experience, analysis, or solid theoretical work and will relate to the workshop problem or an aspect of the problem. Papers should be about one page in length, or two if there are diagrams, written in a format that can be rendered by MS Word. Depending on the number of responses, the workshop committee may take the papers into consideration in the selection of participants. Participants will be notified by e-mail no later than July 24. This is to be a group effort, and all contributions, spoken or written, become property of the IEEE and may be used in IEEE workshop reports without specific attribution. Workshop Program Committee Bob Coyne, IBM Software Group Dick Watson, Lawrence Livermore National Laboratory Reagan Moore, San Diego Supercomputer Center Paul Rutherford, Isilon Systems Harry Hulen, IBM Global Business Services -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 24384066.gif Type: image/gif Size: 4314 bytes Desc: not available URL: From ORODEH at il.ibm.com Tue May 8 01:56:29 2007 From: ORODEH at il.ibm.com (Ohad Rodeh) Date: Tue, 8 May 2007 08:56:29 +0300 Subject: [Storage-research-list] Looking for the cello trace Message-ID: Hello, I am looking to do some research on file-system traces. Can anybody tell me where I can download large-scale traces? For example, the cello trace from HP? Thanks, Ohad. From Mark.Carlson at Sun.COM Tue May 8 10:03:52 2007 From: Mark.Carlson at Sun.COM (Mark A. Carlson) Date: Tue, 08 May 2007 08:03:52 -0600 Subject: [Storage-research-list] Looking for the cello trace In-Reply-To: References: Message-ID: <46408348.5040303@sun.com> Check out: http://iotta.snia.org/ -- mark Ohad Rodeh wrote: > Hello, > I am looking to do some research on file-system traces. Can anybody tell > me where I can download large-scale traces? For example, the cello trace > from HP? > > Thanks, > Ohad. > > _______________________________________________ > Storage-research-list mailing list > Storage-research-list at ece.cmu.edu > https://sos.ece.cmu.edu/mailman/listinfo/storage-research-list > From andrew5 at ece.cmu.edu Tue May 8 10:09:41 2007 From: andrew5 at ece.cmu.edu (Andrew Klosterman) Date: Tue, 8 May 2007 10:09:41 -0400 (EDT) Subject: [Storage-research-list] Looking for the cello trace In-Reply-To: References: Message-ID: SNIA has an Input/Output Traces, Tools, and Analysis Technical Working Group. They have a collection of traces (including cello traces) at the following URL... http://iotta.snia.org/traces/ On Tue, 8 May 2007, Ohad Rodeh wrote: > > Hello, > I am looking to do some research on file-system traces. Can anybody tell > me where I can download large-scale traces? For example, the cello trace > from HP? > > Thanks, > Ohad. > > _______________________________________________ > Storage-research-list mailing list > Storage-research-list at ece.cmu.edu > https://sos.ece.cmu.edu/mailman/listinfo/storage-research-list > --Andrew J. Klosterman andrew5 at ece.cmu.edu From ORODEH at il.ibm.com Tue May 8 10:22:31 2007 From: ORODEH at il.ibm.com (Ohad Rodeh) Date: Tue, 8 May 2007 17:22:31 +0300 Subject: [Storage-research-list] File-system traces Message-ID: List, Thanks for all the responses I got. I had made a mistake in asking for the cello traces, I actually need a file-system, not block-level, trace. I stand corrected. Anyhow, I have one comment, the SNIA repository at http://iotta.snia.org/traces/ does not contain any NFS traces currently. I hope that will change in the future. Ohad. From atraeger at cs.sunysb.edu Tue May 8 12:05:58 2007 From: atraeger at cs.sunysb.edu (Avishay Traeger) Date: Tue, 08 May 2007 12:05:58 -0400 Subject: [Storage-research-list] Looking for the cello trace Message-ID: <1178640358.20855.1.camel@rockstar.fsl.cs.sunysb.edu> Ohad, It looks like IOTTA only has the 1991, 1992, and 1996 Cello traces. HP has the 1999 trace (as well as others) available at: http://tesla.hpl.hp.com/public_software/ Avishay From ganger at ece.cmu.edu Wed May 9 10:38:09 2007 From: ganger at ece.cmu.edu (Greg Ganger) Date: Wed, 9 May 2007 10:38:09 -0400 (EDT) Subject: [Storage-research-list] File-system traces In-Reply-To: References: Message-ID: Of course, many of us share your hope. Please note that IBM is absolutely welcome to contribute NFS traces (or any other I/O traces) to this community repository. Greg On Tue, 8 May 2007, Ohad Rodeh wrote: > > List, > Thanks for all the responses I got. I had made a mistake in asking for > the cello traces, I actually need a file-system, not block-level, trace. I > stand corrected. Anyhow, I have one comment, the SNIA repository at > http://iotta.snia.org/traces/ does not contain any NFS traces currently. I > hope that will change in the future. > > Ohad. > > _______________________________________________ > Storage-research-list mailing list > Storage-research-list at ece.cmu.edu > https://sos.ece.cmu.edu/mailman/listinfo/storage-research-list > From alistair.veitch at hp.com Wed May 9 13:16:46 2007 From: alistair.veitch at hp.com (Alistair Veitch) Date: Wed, 09 May 2007 10:16:46 -0700 Subject: [Storage-research-list] Looking for the cello trace In-Reply-To: <1178640358.20855.1.camel@rockstar.fsl.cs.sunysb.edu> References: <1178640358.20855.1.camel@rockstar.fsl.cs.sunysb.edu> Message-ID: <464201FE.7080707@hp.com> With regards to cello traces, we (HP) hope to have full-year 2003 traces available later this month. IOTTA will have these as well (BTW, IOTTA *do* have the 1999 traces, the website is just undergoing some work to cope with the number/size of files in that set). With regards to NFS traces, we hope to have news there soon as well. Alistair Avishay Traeger wrote: > Ohad, > It looks like IOTTA only has the 1991, 1992, and 1996 Cello traces. HP > has the 1999 trace (as well as others) available at: > > http://tesla.hpl.hp.com/public_software/ > > Avishay > > _______________________________________________ > Storage-research-list mailing list > Storage-research-list at ece.cmu.edu > https://sos.ece.cmu.edu/mailman/listinfo/storage-research-list > From chase at cs.duke.edu Wed May 9 13:34:08 2007 From: chase at cs.duke.edu (Jeff Chase) Date: Wed, 09 May 2007 13:34:08 -0400 Subject: [Storage-research-list] File-system traces In-Reply-To: References: Message-ID: <46420610.1090301@cs.duke.edu> Forgive me if this is off-topic, but it seems like a good time and place to put in a plug for Darrell Anderson's fstress package, which is a synthetic trace generator and load generator for NFSv3/UDP. It generates NFS traces from a set of input parameters that define reference distributions etc. etc. If your goal is to learn something from the traces, then it won't help you. But if your goal is to evaluate storage servers then it can be very helpful. http://www.cs.duke.edu/ari/fstress/ Fstress has canned settings for several "representative" NFS workloads, including SPECsfs. If and when a larger set of NFS traces does become available, it would be useful to extract the parameters to generate a "similar" workload synthetically, e.g., with fstress. Of course this is a research topic. Fstress has been quietly aging, but we have recently started using it again and it seems to be pretty robust. A good group of folks have been using in industry, but I still see some papers go by from the research community with really hoary evaluation methodologies like "the Andrew benchmarks" or much more limited workload gens like Postmark. As a community I think we'd be better off with evaluations based on fstress until something better comes along. Jeff Greg Ganger wrote: > > Of course, many of us share your hope. Please note that IBM is > absolutely > welcome to contribute NFS traces (or any other I/O traces) to this > community > repository. > > Greg > > > On Tue, 8 May 2007, Ohad Rodeh wrote: > >> >> List, >> Thanks for all the responses I got. I had made a mistake in asking for >> the cello traces, I actually need a file-system, not block-level, >> trace. I >> stand corrected. Anyhow, I have one comment, the SNIA repository at >> http://iotta.snia.org/traces/ does not contain any NFS traces >> currently. I >> hope that will change in the future. >> >> Ohad. >> >> _______________________________________________ >> Storage-research-list mailing list >> Storage-research-list at ece.cmu.edu >> https://sos.ece.cmu.edu/mailman/listinfo/storage-research-list >> > _______________________________________________ > Storage-research-list mailing list > Storage-research-list at ece.cmu.edu > https://sos.ece.cmu.edu/mailman/listinfo/storage-research-list From ezk at cs.sunysb.edu Wed May 9 21:18:22 2007 From: ezk at cs.sunysb.edu (Erez Zadok) Date: Wed, 9 May 2007 21:18:22 -0400 Subject: [Storage-research-list] File-system traces In-Reply-To: Your message of "Tue, 08 May 2007 17:22:31 +0300." Message-ID: <200705100118.l4A1IMC4030213@agora.fsl.cs.sunysb.edu> In message , Ohad Rodeh writes: > > List, > Thanks for all the responses I got. I had made a mistake in asking for > the cello traces, I actually need a file-system, not block-level, trace. I > stand corrected. Anyhow, I have one comment, the SNIA repository at > http://iotta.snia.org/traces/ does not contain any NFS traces currently. I > hope that will change in the future. > > Ohad. SNIA has a few of our small f/s level traces captured using Tracefs and replayable using Replayfs. The tools were released too, so you could even capture and replay your own traces. In addition to SNIA, you can find our tools and traces here: Erez. From simsong at acm.org Wed May 9 16:40:45 2007 From: simsong at acm.org (Simson Garfinkel) Date: Wed, 09 May 2007 16:40:45 -0400 Subject: [Storage-research-list] File-system traces In-Reply-To: <46420610.1090301@cs.duke.edu> References: <46420610.1090301@cs.duke.edu> Message-ID: <464231CD.6070600@acm.org> Perhaps even more off-topic, but related nonetheless. I have a page of "forensic corpora" on the Forensics Wiki, at: http://www.forensicswiki.org/wiki/Forensic_corpora Jeff Chase wrote: > Forgive me if this is off-topic, but it seems like a good time and > place to put in a plug for Darrell Anderson's fstress package, which > is a synthetic trace generator and load generator for NFSv3/UDP. It > generates NFS traces from a set of input parameters that define > reference distributions etc. etc. If your goal is to learn something > from the traces, then it won't help you. But if your goal is to > evaluate storage servers then it can be very helpful. > http://www.cs.duke.edu/ari/fstress/ > > Fstress has canned settings for several "representative" NFS > workloads, including SPECsfs. If and when a larger set of NFS traces > does become available, it would be useful to extract the parameters to > generate a "similar" workload synthetically, e.g., with fstress. Of > course this is a research topic. > > Fstress has been quietly aging, but we have recently started using it > again and it seems to be pretty robust. A good group of folks have > been using in industry, but I still see some papers go by from the > research community with really hoary evaluation methodologies like > "the Andrew benchmarks" or much more limited workload gens like > Postmark. As a community I think we'd be better off with evaluations > based on fstress until something better comes along. > > Jeff > > Greg Ganger wrote: >> >> Of course, many of us share your hope. Please note that IBM is >> absolutely >> welcome to contribute NFS traces (or any other I/O traces) to this >> community >> repository. >> >> Greg >> >> >> On Tue, 8 May 2007, Ohad Rodeh wrote: >> >>> >>> List, >>> Thanks for all the responses I got. I had made a mistake in asking >>> for >>> the cello traces, I actually need a file-system, not block-level, >>> trace. I >>> stand corrected. Anyhow, I have one comment, the SNIA repository at >>> http://iotta.snia.org/traces/ does not contain any NFS traces >>> currently. I >>> hope that will change in the future. >>> >>> Ohad. >>> >>> _______________________________________________ >>> Storage-research-list mailing list >>> Storage-research-list at ece.cmu.edu >>> https://sos.ece.cmu.edu/mailman/listinfo/storage-research-list >>> >> _______________________________________________ >> Storage-research-list mailing list >> Storage-research-list at ece.cmu.edu >> https://sos.ece.cmu.edu/mailman/listinfo/storage-research-list > > _______________________________________________ > Storage-research-list mailing list > Storage-research-list at ece.cmu.edu > https://sos.ece.cmu.edu/mailman/listinfo/storage-research-list > From tdumitra at ece.cmu.edu Thu May 10 21:01:33 2007 From: tdumitra at ece.cmu.edu (Tudor Dumitras) Date: Thu, 10 May 2007 21:01:33 -0400 Subject: [Storage-research-list] 1st CFP: ICSOC07 Ph.D. Symposium Message-ID: <4643C06D.3080102@ece.cmu.edu> We apologize if you receive multiple copies of this CfP. CALL FOR PAPERS IBM Ph.D. Symposium at the International Conference on Service-Oriented Computing Vienna, Austria September 17, 2007 http://infolab.uvt.nl/phd-icsoc07/ Following the past successes of the Ph.D. Symposia in 2005 and 2006, the 3rd IBM Ph.D. Symposium on Service-Oriented Computing will take place in Vienna, in conjunction with the International Conference in Service-Oriented Computing (ICSOC 2007). OBJECTIVES The ICSOC Ph.D. Symposium is focused on mentoring doctoral students who are close to finishing their dissertations. Participating students will present their work in front of a mock thesis committee of 4-5 senior researchers in the field who will act as mentors and will provide extensive feedback and advice for preparing a successful Ph.D. dissertation. Participants will also have the opportunity to ask questions about research careers in industry and academia during a panel discussion. The goals of the Symposium are to expose students to constructive criticism before their thesis defense, to serve as an opportunity for networking with other students at a similar stage in their careers, and to provide guidance related to future career perspectives. SCOPE The Symposium has a similar technical scope to ICSOC. We seek original papers in the field of service-oriented computing, from theoretical and foundational results to empirical evaluations as well as practical and industrial experiences, with the emphasis on results that contribute to solve the many still open research problems that are of significant impact to the field of service oriented applications. Topics include but are not limited to the following: * Business Service Modeling: Methods and tools for capturing business goals and requirements, decomposition into business services, business processes, business policies, modeling, analysis, and simulation, specification of functional and non-functional quality requirements * Service Assembly: Development and discovery, model-driven development, service composition architectures, service registries, service discovery mechanisms, semantic matching, methods and tools for service development, governance, verification and validation, deployment strategies * Service Management: Instrumentation and service-related data aggregation, end-to-end measurement, analysis, modeling and capacity planning, definition of deployment topology, infrastructure configuration, problem determination for SOAs, ITIL processes, change management in live systems * SOA Runtime: Service bus for mediation, transformation and routing, runtime development and service registries, integration of legacy applications, information services for data access and data integration, scalability, topology and optimization, service-oriented middleware, policy based configuration & workload management * Quality of Service: Reliable service-oriented computing, security and privacy in service-oriented computing, SLA and policy specification, QoS negotiation, autonomic management of service levels, empirical studies and benchmarking of QoS, performance and dependability prediction in SOA * Grid Services: Services and architecture for management of infrastructural resources, data and compute intensive applications, execution and resource allocation services for job scheduling, protocols for coordination across multiple resource managers, business-value based allocation, innovative strategies for creation and management of virtual enterprises and organizations, prototype systems and toolkits SUBMISSIONS Each submission must have a Ph.D. student as the sole author. The student?s research must be advanced enough to constitute a concrete research proposal with some preliminary results, and the student should be interested in receiving constructive feedback on his/her Ph.D. dissertation. Ideal participants are typically, but not exclusively, one year away from completing their theses. The submission should highlight the novel ideas of the author?s Ph.D. dissertation. The author?s contributions should be framed in the context of related work (previous approaches, relevant standards, etc.), emphasizing the deficiencies and limitations of the state-of-the-art, as well as the author?s proposed strategies for addressing these issues. Each submission will be reviewed by at least three members of the Ph.D. Symposium technical program committee. The main evaluation criteria are: the maturity of the dissertation research, the quality of the research, the potential for impact, and the relevance to service-oriented computing. Submissions are strictly limited to six pages, following the Lecture Notes in Computer Science format from Springer-Verlag (http://www.springer.de/comp/lncs/authors.html). The proceedings of the workshop will be published online as an IBM research report and possibly in other online forums. Papers can be submitted online at http://infolab.uvt.nl/phd-icsoc07/. TRAVEL GRANTS A limited number of travel grants will be awarded to students whose papers are accepted for the Symposium; the grant will cover part of the travel costs for the student to attend the Symposium. Details will be announced at the Symposium website. KEYNOTE Dr. Paolo Traverso will deliver a keynote speech titled ?Service-Oriented Computing from Design-Time to Run-Time: Some Research Challenges?. The keynote will identify several open questions related to the shift to a service-oriented computing paradigm, providing many opportunities for high-impact Ph.D. research. Dr. Traverso is director of research at the Centro per la ricerca scientifica e tecnologica (IRST) in Trento, Italy, where he leads a division working on software and services, knowledge management and embedded systems. His career included positions in academia and industry. He has served on the editorial board of several journals and he has been the General and Program Chair of ICSOC in 2004 and 2005. IMPORTANT DATES Submission deadline: July 9, 2007 Notification of acceptance: August 12, 2007 Camera-ready submission deadline: August 26, 2007 Symposium: September 17, 2007 SYMPOSIUM ORGANIZATION Symposium Chairs * Andreas Hanemann (Leibniz Supercomputing Center, DE) * Benedikt Kratz (Tilburg University, NL) * Tudor Dumitras (Carnegie Mellon University, USA) * Jyotishman Pathak (Iowa State University, USA) (Preliminary) Technical Program Committee * Claudio Bartolini (HP Labs, USA) * Elisabetta di Nitto (Politecnico di Milano, IT) * Rik Eshuis (University of Eindhoven, NL) * Frank Leymann (University of Stuttgart, DE) * Priya Narasimhan (Carnegie Mellon University, USA) * Daniela Rosu (IBM Research, USA) * Jennifer Schopf (Argonne National Lab, USA) * George Spanoudakis (City University London, UK) * Paolo Traverso (ITC/IRST, IT) * Petr Tuma (Charles University, Prague, CZ) * Willem-Jan van den Heuvel (Tilburg University, NL) (Preliminary) Mentoring Committee * Priya Narasimhan * Paolo Traverso * Petr Tuma * Willem-Jan van den Heuvel -- ______________________________________ Tudor A. Dumitras ECE Department Carnegie Mellon University http://www.ece.cmu.edu/~tdumitra From zhu at eece.maine.edu Mon May 14 10:17:57 2007 From: zhu at eece.maine.edu (Yifeng Zhu) Date: Mon, 14 May 2007 10:17:57 -0400 (EDT) Subject: [Storage-research-list] CFP: SNAPI 2007 Message-ID: [Apologies if you receive this invitation more than once] **************************************************************************** * The 4th International Workshop on Storage Network * Architecture and Parallel I/Os (SNAPI) * September 24, 2007 (Monday), San Diego, California, USA * In conjunction with the 24th IEEE Conference on Mass Storage * Systems and Technologies (MSST'07) * * Website: * http://www.eece.maine.edu/snapi/ * Important dates: * Paper submission: June 20 * Notification of acceptance: July 20 * Final Camera-ready paper: August 20 **************************************************************************** Data are the life-blood of computing and the main asset of any organization. Therefore, disk I/O and data storage on which data reside are becoming "first class citizens" in today's information world. This workshop intends to bring together researchers and practitioners from academia and industry to discuss cutting edge research on parallel and distributed data storage technologies. By discussing ongoing research, the workshop will expose participants to the most recent developments in storage network architectures and parallel I/O. Topics of interest include but are not limited to: * Networked Storage Manageability, Reliability, and Availability * Networked Storage Performance and Scalability * File systems, Object-based storage, Block-level storage * NAS and SAN architectures * Storage networking: e.g. Fibre Channel, InfiniBand, IP Storage, iSCSI * Parallel I/O architectures * Caching and consistency * Evaluation of networked storage architectures * Storage management software * Distributed metadata management Authors are invited to submit an extended abstract no longer than 4 pages for consideration. Submissions should be viewable by Adobe Acrobat Reader (version 3.0 or higher). Accepted papers must be no longer than 8 single-spaced pages (including figures, references, and appendices) using 12pt font. All accepted papers will be presented at the workshop, included in a workshop proceedings booklet (distributed at the workshop), and made available online. Steering Chairs: Qing Yang, University of Rhode Island Hong Jiang, University of Nebraska-Lincoln General Chairs: Xubin He, Tennessee Tech University Yifeng Zhu, University of Maine IEEE Sponsor: Merritt Jones, IEEE Technical Chair for Mass Storage Program Committees: Ahmed Amer, University of Pittsburgh, USA Roger D. Chamberlain, Washington University, USA Phillip M. Dickens, University of Maine, USA Christian Engelmann, Oak Ridge National Lab, USA Dan Feng, Huazhong University of Science and Technology, China Jizhong Han, Chinese Academy of Sciences, China Young-Sik Jeong, Wonkwang University, Korea David R. Kaeli, Northeastern University, USA Xueming Li, Qongqing University, China Ben Kobler, NASA Goddard Space Flight Center, USA Beniamino Di Martino, Second Univ. of Naples, Italy Geyong Min, Bradford University, UK, Li Ou, Scalable Systems Group, Dell Inc, USA Dhabaleswar K. Panda, Ohio State University, USA Jehan-Francois Paris, University of Houston, USA Robert D. Russel, University of New Hampshire, USA Dilma Da Silva, IBM T.J. Watson Research Center, USA Peter Sobe, University of Lubeck, Germany Peter Varman, Rice University, USA Weijun Xiao, University of Rhode Island, USA Tao Xie, San Diego State University, USA Zhiyong Xu, Suffolk University, USA Laurence T. Yang, St. Francis Xavier University, Canada From coyne at us.ibm.com Mon May 14 19:05:56 2007 From: coyne at us.ibm.com (Bob Coyne Jr) Date: Mon, 14 May 2007 18:05:56 -0500 Subject: [Storage-research-list] CFP - IEEE Archive LifeCycle Workshop Message-ID: [Apologies if you receive this invitation more than once] September 24, San Diego, California, USA Archival Storage Life Cycle Management A Workshop at the Twenty-Fourth IEEE Conference on Mass Storage Systems and Technologies Workshop Description The one-day workshop will examine the issues and approaches to providing data stewardship in very large data environments where copying all the data from one generation system to another is not feasible. There are multiple technical, policy, legal and other related life cycle matters that affect data stewardship. Examples are: media aging and reliability characteristics; evolution over time of file system and archive software; standards; government initiatives, programs contracting trends; laws that affect retention management; security policies and technology; and the general evolution of storage technology. These matters will be discussed and expanded in terms of importance and difficulty, and the solution space will be qualified where time and agreement allow. The effort is intended to be a vendor-neutral exercise. The workshop will take place on Monday, September 24, 2007, the day before the beginning of general sessions of the IEEE Mass Storage Systems and Technology Conference MSST2007. See the conference web site: www.storageconference.org/2007 A short white paper is requested from each participant in the workshop, as described below. The white papers will be correlated before the workshop to create a framework for considering approaches and possible solutions. The workshop will generate a report to be published by the IEEE Computer Society Press with a list of participants included, and a summary of the report will be presented during the conference. How to Participate Send email requesting participation to: archivestorage at ieee.org Participants are asked to submit white papers by June 24, 2007. The white paper will be based on experience, analysis, or solid theoretical work and will relate to the workshop problem or an aspect of the problem. Papers should be about one page in length, or two if there are diagrams, written in a format that can be rendered by MS Word. Depending on the number of responses, the workshop committee may take the papers into consideration in the selection of participants. Participants will be notified by e-mail no later than July 24. This is to be a group effort, and all contributions, spoken or written, become property of the IEEE and may be used in IEEE workshop reports without specific attribution. Workshop Program Committee Bob Coyne, IBM Software Group Dick Watson, Lawrence Livermore National Laboratory Hariett Coverston, Sun Microsystems Reagan Moore, San Diego Supercomputer Center Paul Rutherford, Isilon Systems Harry Hulen, IBM Global Business Services -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 26180180.gif Type: image/gif Size: 4314 bytes Desc: not available URL: From Erik.Riedel at seagate.com Wed May 23 09:26:57 2007 From: Erik.Riedel at seagate.com (Erik.Riedel at seagate.com) Date: Wed, 23 May 2007 09:26:57 -0400 Subject: [Storage-research-list] reminder - 5th snia storage security summit - 30, 31 may & 1 june - pittsburgh Message-ID: A reminder of this event happening next week in Pittsburgh. The 5th SNIA Storage Security Summit is on 30, 31 May and 1 June 2007 in Pittsburgh, PA. A day of tutorials for industry professionals will take place on 30 May at Seagate Research followed by a 1.5 day conference at Carnegie Mellon on 31 May and 1 June. The event brings together engineers, architects and product managers from the storage industry building storage security products today and academic researchers looking at future challenges and solutions. We have arranged special discount rates for students interested in current industry topics and thinking. Topics at past summits have covered the range of storage security issues - compliance for managing corporate records; privacy protection for personal data; and the protection of the critical information infrastructure. Professor Carl Landwehr of the University of Maryland (former CyberTrust program manager at NSF) will keynote, along with a range of other speakers on security solutions, usability, legal issues, end user concerns, and future technology directions. Details of the event are online at: http://www.snia.org/security_summit The SNIA Security Summit is held twice annually alternating at UC San Diego and Carnegie Mellon. If anyone is interested in presenting at the January 2008 event in San Diego, please contact me. We hope to see you in Pittsburgh next week. Cheers, Erik writing for the SNIA Storage Security Industry Forum (SSIF) 5th SNIA Storage Security Summit - co-sponsored by SNIA SSIF & Carnegie Mellon University Erik Riedel Department Head Interfaces & Architecture Seagate Research Pittsburgh, PA erik.riedel at seagate.com From zhu at eece.maine.edu Tue Jun 19 21:32:44 2007 From: zhu at eece.maine.edu (Yifeng Zhu) Date: Tue, 19 Jun 2007 21:32:44 -0400 (EDT) Subject: [Storage-research-list] SNAPI 2007 (submission deadline extended to July 1) Message-ID: [Apologies if you receive this invitation more than once] **************************************************************************** * The 4th International Workshop on Storage Network * Architecture and Parallel I/Os (SNAPI) * September 24, 2007 (Monday), San Diego, California, USA * In conjunction with the 24th IEEE Conference on Mass Storage * Systems and Technologies (MSST'07) * Website: * http://www.eece.maine.edu/snapi/ * Important dates: * Paper submission: July 1 * Notification of acceptance: July 20 * Final Camera-ready paper: August 20 **************************************************************************** Data are the life-blood of computing and the main asset of any organization. Therefore, disk I/O and data storage on which data reside are becoming "first class citizens" in today's information world. This workshop intends to bring together researchers and practitioners from academia and industry to discuss cutting edge research on parallel and distributed data storage technologies. By discussing ongoing research, the workshop will expose participants to the most recent developments in storage network architectures and parallel I/O. Topics of interest include but are not limited to: * Networked Storage Manageability, Reliability, and Availability * Networked Storage Performance and Scalability * File systems, Object-based storage, Block-level storage * NAS and SAN architectures * Storage networking: e.g. Fibre Channel, InfiniBand, IP Storage, iSCSI * Parallel I/O architectures * Caching and consistency * Evaluation of networked storage architectures * Storage management software * Distributed metadata management Authors are invited to submit an extended abstract no longer than 4 pages for consideration. Submissions should be viewable by Adobe Acrobat Reader (version 3.0 or higher). Accepted papers must be no longer than 8 single-spaced pages (including figures, references, and appendices) using 12pt font. All accepted papers will be presented at the workshop, included in a workshop proceedings booklet (distributed at the workshop), and made available online. Steering Chairs: Qing Yang, University of Rhode Island Hong Jiang, University of Nebraska-Lincoln General Chairs: Xubin He, Tennessee Tech University Yifeng Zhu, University of Maine IEEE Sponsor: Merritt Jones, IEEE Technical Chair for Mass Storage Program Committees: Ahmed Amer, University of Pittsburgh, USA Roger D. Chamberlain, Washington University, USA Phillip M. Dickens, University of Maine, USA Christian Engelmann, Oak Ridge National Lab, USA Dan Feng, Huazhong University of Science and Technology, China Jizhong Han, Chinese Academy of Sciences, China Young-Sik Jeong, Wonkwang University, Korea David R. Kaeli, Northeastern University, USA Xueming Li, Qongqing University, China Ben Kobler, NASA Goddard Space Flight Center, USA Beniamino Di Martino, Second Univ. of Naples, Italy Geyong Min, Bradford University, UK, Li Ou, Scalable Systems Group, Dell Inc, USA Dhabaleswar K. Panda, Ohio State University, USA Jehan-Francois Paris, University of Houston, USA Robert D. Russel, University of New Hampshire, USA Dilma Da Silva, IBM T.J. Watson Research Center, USA Peter Sobe, University of Lubeck, Germany Peter Varman, Rice University, USA Weijun Xiao, University of Rhode Island, USA Tao Xie, San Diego State University, USA Zhiyong Xu, Suffolk University, USA Laurence T. Yang, St. Francis Xavier University, Canada From rick.bauer at snia.org Tue Aug 7 11:40:23 2007 From: rick.bauer at snia.org (Bauer, Rick) Date: Tue, 7 Aug 2007 09:40:23 -0600 Subject: [Storage-research-list] (no subject) Message-ID: <7FE20DCBBC9CD747A29DF177CD0E002D063EC778@VS6.EXCHPROD.USA.NET> Please post to the reflector...SNIA Storage Developer Conference.....text follows...... Rick Rick Bauer, Technology Director Storage Networking Industry Association www.snia.org telephone: (719) 884-8900 (SNIA Tech Center, US) e-mail: rick.bauer at snia.org SNIA's Storage Developer Conference Where the Storage Development Community Connects San Jose, CA * Dolce Hayes Manion * September 10-13, 2007 New Solutions from the Best Storage Development Minds in Security, Technology and Management Now expanded to include 1 FULL DAY of technical tutorials - with an agenda designed by SNIA's Technical Council, this year's annual conference covers all the storage development issues utmost on your mind right now. You'll hear about File & Object Storage , Advanced Technologies, Storage Management, NFS, Storage Software, iSCSI, Security and more. This is the only conference that focuses on the issues that matter to developers. That's why so many storage application developers, storage management developers, filesystem or host software developers, and storage protocol implementers make it a point to return to this conference every year. Join us September 10-13, 2007 at the Dolce Hayes Mansion in San Jose, CA. SNIA Annual Storage Developer Conference Co-Located with CIFS & iSCSI 2007 Plugfest * 10 Tutorial Sessions * 2 Nationally Recognized Keynote Speakers * 36 In-Depth Education Sessions * 4 General Sessions with Featured Speakers Conference Highlights * Keynote Addresses from Richard Stallman, founder of the GNU Project and Paul Strong, Distinguished Research Scientist, eBay, Inc. * Actionable analysis and insights in hot topic areas such as XAM (eXtensible Access Method), ZFS (Zetabyte File System) and OSD (Object Based Storage Devices). * Featured Speakers - Lawrence Rosen, Rosenlaw & Einschlag. Jeff Bonwick, Sun Microsystems. * Reception and Plugfest Open House - Interactive forum in which companies can develop compatible CIFS and iSCSI products. * Sponsor Showcase - You'll find the products and services you need to transform, upgrade and improve your business. Don't miss this event! Register today for SNIA's Storage Developer Conference at www.storage-developer.org. Register today for SNIA's Storage Developer Conference at http://www.storage-developer.org September 10-13, 2007 San Jose, CA Rick Rick Bauer, Technology Director Storage Networking Industry Association www.snia.org telephone: (719) 884-8900 (SNIA Tech Center, US) e-mail: rick.bauer at snia.org -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: SDC 2007 Flyer.pdf Type: application/octet-stream Size: 125301 bytes Desc: SDC 2007 Flyer.pdf URL: From garth at cs.cmu.edu Sun Sep 9 12:07:11 2007 From: garth at cs.cmu.edu (Garth gibson) Date: Sun, 9 Sep 2007 12:07:11 -0400 Subject: [Storage-research-list] CFP: Petascale Data Storage Workshop at SC07 (Nov 11, 2007) Message-ID: 2007 DOE Office of Science Petascale Data Storage Workshop Int. Conf. for High perfomrance Computing, Networking, Storage and Analysis (SC07) Sunday November 11, 2007, 8:30 am - 5 pm Atlantis Hotel, Ballroom B Reno, Nevada www.pdsi-scidac.org/SC07/ Petascale computing infrastructures make petascale demands on information storage capacity, performance, concurrency, reliability, availability, and manageability. This one-day workshop focuses on the data storage problems and emerging solutions found in petascale scientific computing environments, with special attention to issues in which community collaboration can be crucial, problem identification, workload capture, solution interoperability, standards with community buy-in, and shared tools. Submission: http://www.pdsi-scidac.org/SC07/submit.html Paper (extended abstract in pdf) due Sept. 28, 2007 Notification: Oct. 11, 2007 Final softcopy and slides due Nov. 11 at the workshop This year the petascale data storage workshop will hold a peer reviewed competitive process for selecting extended abstracts and short papers. Submit a not previously published extended abstract of 2 to 5 pages, not less than 10 point font, in a PDF file as instructed on the workshop web site. Submitted papers will be reviewed under the supervision of the workshop program committee. Submissions should indicate authors and affiliations. Selected final papers may be longer than in submission, but not longer than 10 pages. The workshop proceedings will be published in the ACM digital library and talk slides will be made available on the workshop web site. Committee: Garth A. Gibson, Carnegie Mellon University and Panasas Inc. Darrell Long, University of California, Santa Cruz Peter Honeyman, University of Michigan, Ann Arbor, Center for Information Technology Integration Gary A. Grider, Los Alamos National Laboratory William T.C. Kramer, National Energy Research Scientific Computing Center, Lawrence Berkeley National Laboratory Philip C. Roth, Oak Ridge National Laboratory Evan J. Felix, Pacific Northwest National Laboratory Lee Ward, Sandia National Laboratory -------------- next part -------------- A non-text attachment was scrubbed... Name: cfp.pdf Type: application/pdf Size: 610225 bytes Desc: not available URL: From Erik.Riedel at seagate.com Mon Sep 10 09:04:56 2007 From: Erik.Riedel at seagate.com (Erik.Riedel at seagate.com) Date: Mon, 10 Sep 2007 09:04:56 -0400 Subject: [Storage-research-list] reminder - FAST '08 papers due this week Message-ID: FAST '08 Call for Papers Deadline Approaching: September 12, 2007 The 6th USENIX Conference on File and Storage Technologies (FAST '08) Program Committee would like to remind you that the deadline to contribute to the refereed papers is quickly approaching. Paper submissions are due 9:00 p.m. EDT, September 12, 2007. FAST brings together the top storage systems researchers and practitioners, providing the premier forum for discussing the design, implementation, evaluation, and deployment of storage systems. FAST '08 will take place February 26-29, 2008, in San Jose, CA. For topics of interest and submission guidelines, see http://www.usenix.org/fast08/cfpmes From garth at cs.cmu.edu Mon Sep 24 14:53:47 2007 From: garth at cs.cmu.edu (Garth gibson) Date: Mon, 24 Sep 2007 14:53:47 -0400 Subject: [Storage-research-list] (Sept 28 deadline) CFP: Petascale Data Storage Workshop at SC07 (Nov 11, 2007) Message-ID: 2007 DOE Office of Science Petascale Data Storage Workshop Int. Conf. for High perfomrance Computing, Networking, Storage and Analysis (SC07) Sunday November 11, 2007, 8:30 am - 5 pm Atlantis Hotel, Ballroom B Reno, Nevada www.pdsi-scidac.org/SC07/ Petascale computing infrastructures make petascale demands on information storage capacity, performance, concurrency, reliability, availability, and manageability. This one-day workshop focuses on the data storage problems and emerging solutions found in petascale scientific computing environments, with special attention to issues in which community collaboration can be crucial, problem identification, workload capture, solution interoperability, standards with community buy-in, and shared tools. Submission: http://www.pdsi-scidac.org/SC07/submit.html Paper (extended abstract in pdf) due Sept. 28, 2007 Notification: Oct. 11, 2007 Final softcopy and slides due Nov. 11 at the workshop This year the petascale data storage workshop will hold a peer reviewed competitive process for selecting extended abstracts and short papers. Submit a not previously published extended abstract of 2 to 5 pages, not less than 10 point font, in a PDF file as instructed on the workshop web site. Submitted papers will be reviewed under the supervision of the workshop program committee. Submissions should indicate authors and affiliations. Selected final papers may be longer than in submission, but not longer than 10 pages. The workshop proceedings will be published in the ACM digital library and talk slides will be made available on the workshop web site. Committee: Garth A. Gibson, Carnegie Mellon University and Panasas Inc. Darrell Long, University of California, Santa Cruz Peter Honeyman, University of Michigan, Ann Arbor, Center for Information Technology Integration Gary A. Grider, Los Alamos National Laboratory William T.C. Kramer, National Energy Research Scientific Computing Center, Lawrence Berkeley National Laboratory Philip C. Roth, Oak Ridge National Laboratory Evan J. Felix, Pacific Northwest National Laboratory Lee Ward, Sandia National Laboratory ? -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: cfp.pdf Type: application/pdf Size: 610225 bytes Desc: not available URL: -------------- next part -------------- An HTML attachment was scrubbed... URL: From garth at cs.cmu.edu Fri Sep 28 07:46:56 2007 From: garth at cs.cmu.edu (Garth Gibson) Date: Fri, 28 Sep 2007 07:46:56 -0400 Subject: [Storage-research-list] Deadline EXTENDED until 11:59pm PT Sat Sept 29, 2007: Petascale Data Storage Workshop at SC07 (Nov 11, 2007) Message-ID: <5FE0105C-7497-4E40-9C83-4E10B66F97A7@cs.cmu.edu> Submission deadline EXTENDED until 11:59 PM Pacific timezone Saturday September 29, 2007. > From: "Garth gibson" > Date: September 9, 2007 12:07:11 PM EDT (CA) > To: > Subject: [Storage-research-list] CFP: Petascale Data Storage > Workshop at SC07 (Nov 11, 2007) > > 2007 DOE Office of Science Petascale Data Storage Workshop > Int. Conf. for High perfomrance Computing, Networking, Storage and > Analysis (SC07) > Sunday November 11, 2007, 8:30 am - 5 pm > Atlantis Hotel, Ballroom B > Reno, Nevada > www.pdsi-scidac.org/SC07/ > > Petascale computing infrastructures make petascale demands on > information storage capacity, performance, concurrency, reliability, > availability, and manageability. This one-day workshop focuses on the > data storage problems and emerging solutions found in petascale > scientific computing environments, with special attention to issues > in which community collaboration can be crucial, problem > identification, workload capture, solution interoperability, > standards with community buy-in, and shared tools. > > Submission: > http://www.pdsi-scidac.org/SC07/submit.html > Paper (extended abstract in pdf) due Sept. 28, 2007 > Notification: Oct. 11, 2007 > Final softcopy and slides due Nov. 11 at the workshop > > This year the petascale data storage workshop will hold a peer > reviewed competitive process for selecting extended abstracts and > short papers. Submit a not previously published extended abstract of > 2 to 5 pages, not less than 10 point font, in a PDF file as > instructed on the workshop web site. Submitted papers will be > reviewed under the supervision of the workshop program committee. > Submissions should indicate authors and affiliations. Selected final > papers may be longer than in submission, but not longer than 10 > pages. The workshop proceedings will be published in the ACM digital > library and talk slides will be made available on the workshop web > site. > > Committee: > Garth A. Gibson, Carnegie Mellon University and Panasas Inc. > Darrell Long, University of California, Santa Cruz > Peter Honeyman, University of Michigan, Ann Arbor, Center for > Information Technology Integration > Gary A. Grider, Los Alamos National Laboratory > William T.C. Kramer, National Energy Research Scientific Computing > Center, Lawrence Berkeley National Laboratory > Philip C. Roth, Oak Ridge National Laboratory > Evan J. Felix, Pacific Northwest National Laboratory > Lee Ward, Sandia National Laboratory > From andrew5 at ece.cmu.edu Tue Oct 9 08:53:26 2007 From: andrew5 at ece.cmu.edu (Andrew Klosterman) Date: Tue, 9 Oct 2007 08:53:26 -0400 (EDT) Subject: [Storage-research-list] Nobel in Physics for "Giant Magnetoresistance" Message-ID: This year's Nobel Prize in Physics was granted for technology used in modern hard drives. Chalk one up for storage research! http://nobelprize.org/nobel_prizes/physics/laureates/2007/press.html The Royal Swedish Academy of Sciences has decided to award the Nobel Prize in Physics for 2007 jointly to Albert Fert Unite' Mixte de Physique CNRS/THALES, Universit Paris-Sud, Orsay, France, and Peter Gruenberg Forschungszentrum Juelich, Germany, "for the discovery of Giant Magnetoresistance". Nanotechnology gives sensitive read-out heads for compact hard disks This year's physics prize is awarded for the technology that is used to read data on hard disks. It is thanks to this technology that it has been possible to miniaturize hard disks so radically in recent years. Sensitive read-out heads are needed to be able to read data from the compact hard disks used in laptops and some music players, for instance. In 1988 the Frenchman Albert Fert and the German Peter Gruenberg each independently discovered a totally new physical effect -- Giant Magnetoresistance or GMR. Very weak magnetic changes give rise to major differences in electrical resistance in a GMR system. A system of this kind is the perfect tool for reading data from hard disks when information registered magnetically has to be converted to electric current. Soon researchers and engineers began work to enable use of the effect in read-out heads. In 1997 the first read-out head based on the GMR effect was launched and this soon became the standard technology. Even the most recent read-out techniques of today are further developments of GMR. A hard disk stores information, such as music, in the form of microscopically small areas magnetized in different directions. The information is retrieved by a read-out head that scans the disk and registers the magnetic changes. The smaller and more compact the hard disk, the smaller and weaker the individual magnetic areas. More sensitive read-out heads are therefore required if information has to be packed more densely on a hard disk. A read-out head based on the GMR effect can convert very small magnetic changes into differences in electrical resistance and there-fore into changes in the current emitted by the read-out head. The current is the signal from the read-out head and its different strengths represent ones and zeros. The GMR effect was discovered thanks to new techniques developed during the 1970s to produce very thin layers of different materials. If GMR is to work, structures consisting of layers that are only a few atoms thick have to be produced. For this reason GMR can also be considered one of the first real applications of the promising field of nanotechnology. --Andrew J. Klosterman andrew5 at ece.cmu.edu From jwang at cs.ucf.edu Sat Oct 20 23:45:41 2007 From: jwang at cs.ucf.edu (Dr. Jun Wang) Date: Sat, 20 Oct 2007 23:45:41 -0400 (EDT) Subject: [Storage-research-list] CFP for 1st International Workshop on Storage and I/O Virtualization, Performance, Energy, Evaluation and Dependability (SPEED08) Held in conjunction with HPCA2008 Message-ID: Dear colleagues, http://speed2008.eecs.ucf.edu/main.php A new storage and I/O workshop cfp with the focus on hot topics including Virtualization, Performance, Energy, Evaluation and Dependability! Peter and Jun Peter Varman Professor Department of ECE and CS Rice University Houston, TX 77005 Email: pjv at rice.edu Tel: 713-348-3990 Jun Wang, Assistant Professor School of Electrical Engineering and Computer Science University of Central Florida Orlando, FL 32816-2450 Phone: (407) 823-0449 Email: jwang at eecs.ucf.edu Fax: (407) 823-5835 URL: http://www.eecs.ucf.edu/~jwang From eno at andrew.cmu.edu Fri Oct 26 06:37:43 2007 From: eno at andrew.cmu.edu (Eno Thereska) Date: Fri, 26 Oct 2007 06:37:43 -0400 Subject: [Storage-research-list] Flash, Mram, MEMS - what are the trends? Message-ID: <4721C377.5040604@andrew.cmu.edu> Hello, Are there good pointers you'd recommend on where the Flash, Mram, MEMS, (i.e., alternative technologies to disks) are headed? Also, I am interested in best practices for using any of the above (most likely Flash, the others are still in the design phase I suspect) in current storage systems. Any pointers are appreciated. Thanks, Eno Thereska From rick at fsl.cs.sunysb.edu Sun Oct 28 23:01:43 2007 From: rick at fsl.cs.sunysb.edu (Richard Paul Spillane) Date: Sun, 28 Oct 2007 23:01:43 -0400 Subject: [Storage-research-list] Flash, Mram, MEMS - what are the trends? In-Reply-To: <4721C377.5040604@andrew.cmu.edu> References: <4721C377.5040604@andrew.cmu.edu> Message-ID: <20071029030143.GC11215@filer.fsl.cs.sunysb.edu> On Fri, Oct 26, 2007 at 06:37:43AM -0400, Eno Thereska wrote: > Hello, > > Are there good pointers you'd recommend on where the Flash, Mram, MEMS, > (i.e., alternative technologies to disks) are headed? Also, I am > interested in best practices for using any of the above (most likely > Flash, the others are still in the design phase I suspect) in current > storage systems. Any pointers are appreciated. I too am interested. > > Thanks, > Eno Thereska -- Richard P. Spillane Research Assistant File Systems and Storage Lab Stony Brook University rick at fsl.cs.sunysb.edu From seattleplus at gmail.com Mon Oct 29 12:52:47 2007 From: seattleplus at gmail.com (dean hildebrand) Date: Mon, 29 Oct 2007 09:52:47 -0700 Subject: [Storage-research-list] Flash, Mram, MEMS - what are the trends? In-Reply-To: <20071029030143.GC11215@filer.fsl.cs.sunysb.edu> References: <4721C377.5040604@andrew.cmu.edu> <20071029030143.GC11215@filer.fsl.cs.sunysb.edu> Message-ID: http://www.storagemarkets.com/ provides some interesting predictions on storage. I haven't looked through them too closely to see if they are bunk or not.... Dean On 10/28/07, Richard Paul Spillane wrote: > On Fri, Oct 26, 2007 at 06:37:43AM -0400, Eno Thereska wrote: > > Hello, > > > > Are there good pointers you'd recommend on where the Flash, Mram, MEMS, > > (i.e., alternative technologies to disks) are headed? Also, I am > > interested in best practices for using any of the above (most likely > > Flash, the others are still in the design phase I suspect) in current > > storage systems. Any pointers are appreciated. > > I too am interested. > > > > > Thanks, > > Eno Thereska > > -- > Richard P. Spillane > Research Assistant > File Systems and Storage Lab > Stony Brook University > rick at fsl.cs.sunysb.edu > _______________________________________________ > Storage-research-list mailing list > Storage-research-list at ece.cmu.edu > https://sos.ece.cmu.edu/mailman/listinfo/storage-research-list > From jack.cole at ieee.org Mon Oct 29 20:49:25 2007 From: jack.cole at ieee.org (Jack Cole) Date: Mon, 29 Oct 2007 20:49:25 -0400 Subject: [Storage-research-list] Flash, Mram, MEMS - what are the trends? In-Reply-To: <20071029030143.GC11215@filer.fsl.cs.sunysb.edu> References: <4721C377.5040604@andrew.cmu.edu> <20071029030143.GC11215@filer.fsl.cs.sunysb.edu> Message-ID: <596648080710291749y36753e2flc4813c02b50ec52a@mail.gmail.com> wrt best practices, you might be interested in the IEEE standard published in June that concerns authentication in host attachment of transient storage devices now and expanded capabilities of such devices in the near future here is a link to an article about the standard followed by the tinyurl version http://www.computer.org/portal/site/computer/index.jsp?pageID=computer_level1_article&TheCat=1060&path=computer/homepage/April07&file=security.xml&xsl=article.xsl http://tinyurl.com/234l92 On 10/28/07, Richard Paul Spillane wrote: > On Fri, Oct 26, 2007 at 06:37:43AM -0400, Eno Thereska wrote: > > Hello, > > > > Are there good pointers you'd recommend on where the Flash, Mram, MEMS, > > (i.e., alternative technologies to disks) are headed? Also, I am > > interested in best practices for using any of the above (most likely > > Flash, the others are still in the design phase I suspect) in current > > storage systems. Any pointers are appreciated. > > I too am interested. > > > > > Thanks, > > Eno Thereska > > -- > Richard P. Spillane > Research Assistant > File Systems and Storage Lab > Stony Brook University > rick at fsl.cs.sunysb.edu > _______________________________________________ > Storage-research-list mailing list > Storage-research-list at ece.cmu.edu > https://sos.ece.cmu.edu/mailman/listinfo/storage-research-list > From raysmile at gmail.com Wed Oct 31 13:58:19 2007 From: raysmile at gmail.com (Lei Tian) Date: Wed, 31 Oct 2007 10:58:19 -0700 Subject: [Storage-research-list] Are there any block-level trace-driven benchmarks available on the Linux 2.6 platform? Message-ID: <200710311058187133896@gmail.com> Hi all, I am conducting some research on the block-level storage systems and want to evaluate the performance of our approach using some trace-driven benchmarks. Anybody can provide some information where I can get the block-level trace-driven benchmarks available on the linux 2.6 platform? Thank you in advance! Lei Lei Tian 2007-10-31 -------------- next part -------------- An HTML attachment was scrubbed... URL: From geoff at cs.hmc.edu Thu Nov 1 18:36:09 2007 From: geoff at cs.hmc.edu (Geoff Kuenning) Date: 01 Nov 2007 15:36:09 -0700 Subject: [Storage-research-list] Are there any block-level trace-driven benchmarks available on the Linux 2.6 platform? In-Reply-To: <200710311058187133896@gmail.com> References: <200710311058187133896@gmail.com> Message-ID: > I am conducting some research on the block-level storage systems and > want to evaluate the performance of our approach using some > trace-driven benchmarks. Anybody can provide some information where > I can get the block-level trace-driven benchmarks available on the > linux 2.6 platform? Yes, there are block-level traces available at http://iotta.snia.org. -- Geoff Kuenning geoff at cs.hmc.edu http://www.cs.hmc.edu/~geoff/ If it's visually ugly, it's almost certainly a bad design. From raysmile at gmail.com Thu Nov 1 13:12:12 2007 From: raysmile at gmail.com (ray smile) Date: Thu, 1 Nov 2007 11:12:12 -0600 Subject: [Storage-research-list] Are there any block-level trace-drivenbenchmarks available on the Linux 2.6 platform? In-Reply-To: References: <200710311058187133896@gmail.com> Message-ID: Thank you for your information. As far as I know, there is not available block-level trace-replay tools on the website( iotta.snia.org ). However, it seems there are two appropriate benchmark software. The first one is Buttress of HP, and the second one is SPEK of URI. I will contact with them soon. Thank you again! 2007/11/1, Ellard, Daniel : > > Have you looked at the snia.org web site? They are building a repository > of traces. I don't know off-hand if any of them are exactly what you're > looking for (but if you find what you're looking for, and it's NOT on the > snia site, you might want to try to arrange for them to be...). > > Good hunting, > -Dan > > ------------------------------ > *From:* Lei Tian [mailto:raysmile at gmail.com] > *Sent:* Wednesday, October 31, 2007 1:58 PM > *To:* storage-research-list > *Subject:* [Storage-research-list] Are there any block-level > trace-drivenbenchmarks available on the Linux 2.6 platform? > > Hi all, > > I am conducting some research on the block-level storage systems and want > to evaluate the performance of our approach using some trace-driven > benchmarks. Anybody can provide some information where I can get > the block-level trace-driven benchmarks available on the linux 2.6platform? > > Thank you in advance! > > Lei > ------------------------------ > Lei Tian > 2007-10-31 > > -- ----------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: From raysmile at gmail.com Fri Nov 2 23:42:20 2007 From: raysmile at gmail.com (ray smile) Date: Fri, 2 Nov 2007 22:42:20 -0500 Subject: [Storage-research-list] Are there any block-level trace-driven benchmarks available on the Linux 2.6 platform? In-Reply-To: References: <200710311058187133896@gmail.com> Message-ID: Thank you for your information. One of the authors of SPEK, told me that , there is a software tool can do the work I need. That is, blktrace, a block-level trace and trace-replay software. However, it has its own trace format, other formats of the traces can be not directly replayed. On 01 Nov 2007 15:36:09 -0700, Geoff Kuenning wrote: > > I am conducting some research on the block-level storage systems and > > want to evaluate the performance of our approach using some > > trace-driven benchmarks. Anybody can provide some information where > > I can get the block-level trace-driven benchmarks available on the > > linux 2.6 platform? > > Yes, there are block-level traces available at http://iotta.snia.org. > -- > Geoff Kuenning geoff at cs.hmc.edu http://www.cs.hmc.edu/~geoff/ > > If it's visually ugly, it's almost certainly a bad design. > -- ----------------------------------- From ggrider at lanl.gov Mon Nov 19 10:25:15 2007 From: ggrider at lanl.gov (Gary Grider) Date: Mon, 19 Nov 2007 08:25:15 -0700 Subject: [Storage-research-list] Fwd: Georgia Tech Computational Science and Engineering Faculty Positions Message-ID: <6.2.3.4.2.20071119082318.035ed690@cic-mail.lanl.gov> >X-Sieve: CMU Sieve 2.2 >Date: Sun, 18 Nov 2007 19:11:56 -0500 (EST) >From: bader at cc.gatech.edu ("David A. Bader") >To: ggrider at lanl.gov >Subject: Georgia Tech Computational Science and Engineering Faculty >Positions >X-Proofpoint-Virus-Version: vendor=fsecure >engine=4.65.5502:2.3.11,1.2.37,4.0.164 >definitions=2007-11-18_03:2007-11-17,2007-11-18,2007-11-19 signatures=0 >X-Proofpoint-Spam: 0 >X-CTN-5-MailScanner-Information: Please see >http://network.lanl.gov/email/virus-scan.php >X-CTN-5-MailScanner: Found to be clean >X-CTN-5-MailScanner-From: bader at cc.gatech.edu >X-Spam-Status: No > > >FACULTY POSITIONS > >Georgia Institute of Technology >College of Computing >Computational Science and Engineering Division > >The Computational Science and Engineering division within the >College of Computing at the Georgia Institute of Technology invites >applications for tenure-track faculty positions. Applications at all >levels of service will be considered. Applicants must have an >outstanding record of research, a sincere commitment to teaching, >and interest in engaging in substantive interdisciplinary research >with collaborators in other disciplines. We encourage applications >from any areas of computational science and engineering. Applicants >with expertise in high-performance computing (HPC), modeling, >simulation and numerical computing, bioinformatics and computational >biology, and large-scale data analysis and visualization are >especially encouraged to apply. > > >To receive full consideration, applications should be submitted >online through http://recruiting.cc.gatech.edu/ by January 11, 2008. >The application material should include a full academic CV, teaching >and research statements, a list of at least three references and up >to three publications. Applicants are encouraged to clearly >identify in their cover letter the area(s) that best describe their >research interests. > > >Georgia Tech is an Affirmative Action/Equal Opportunity Employer. > >Applications from women and under-represented minorities are >strongly encouraged. > >________________________ >David A. Bader >Executive Director of High-Performance Computing >College of Computing >Georgia Institute of Technology >Atlanta, GA 30332 USA >http://www.cc.gatech.edu/~bader >Assistant: Carolyn Young, cyoung at cc.gatech.edu, 404-385-3425