- FAST '13 Home
- Registration Information
- Registration Discounts
- At a Glance
- Training Program
- Technical Sessions
- Purchase the Box Set
- Posters and WiPs
- Birds-of-a-Feather Sessions
- Hotel and Travel Information
- Help Promote
- For Participants
- Call for Papers
- Past Proceedings
You are here
Scheduling a BoF
To schedule a BoF, simply write the BoF title as well as your name and affiliation on one of the BoF Boards located in the registration area. If you have a description of our BoF you'd like posted on this Web page, please schedule your BoF on the BoF board, then send its title, the organizer's name and affiliation, and the date, time, and location of the BoF to email@example.com with "FAST '13 BoF" in the subject line.
|Tuesday, February 12, 2013
|Wednesday, February 13, 2013
|Tracing and Measurement BoF
|Availability of NSF PRObE Cluster Resources BoF
|Linux File & Storage Update BoF
| NFSv4.1 and pNFS Ready for Prime Time Deployment BoF
|EMC Backup Recovery Systems Hospitality Vendor BoF
|USENIX WiAC: Women in Advanced Computing BoF
|Non-Volatile Memory Programming BoF
|Thursday, February 14, 2013
|VMware Vendor BoF: Software Defined Storage — Challenges and Open Issues
Tracing and Measurement BoF
Wednesday, February 13, 7:30 p.m.-8:30 p.m., Garden
Do you believe in measuring or benchmarking file systems? Are you interested in collecting an analyzing traces? At this BOF we will discuss tracing and measurement techniques, and ways to make the task easier. There will be a brief status update on the SNIA IOTTA Trace Repository, and an announcement of early results from the SNIA SSSI Workload I/O Capture Project.
Daniel Phillips, tux3.org
Wednesday, February 13, 7:30 p.m.-8:30 p.m., Empire
In addition to Tux3 specifics, plans and progress, we will touch on filesystem design in general, and compare to the incumbents: Ext4, ZFS, WAFL. And to the new projects: Btrfs, Hammer2.
Hot Tux3 topics at the moment:
* The road to online and incremental fsck
* High performance atomic commit technology
* Stronger consistency semantics
* Optimizing for both spinning and flash media
* New scalable directory index
* Snapshotting design
Come join members of EMC BRS for snacks and refreshments, and to learn about the interesting technical challenges facing the industry as CTO Stephen Manley discusses where backup is heading. EMC is hiring at all levels, for both full-time positions and internships. Door prize of an iPad mini will be given (must be present to win).
Emerging Non-Volatile Memory (NVM) technologies enable new features that straddle the boundary between storage and memory. The SNIA NVM Programming Technical Workgroup has been researching and defining new programming models for NVM and this BOF will begin with a quick overview of the current progress in this area. Attendees will be encouraged to discuss their expectations for NVM programming and, time permitting, some of the open questions in this area will be discussed as a group.
Availability of NSF PRObE Cluster Resources
Garth Gibson, CMU, and Andree Jacobson, New Mexico Consortium
Wednesday, February 13, 8:30 p.m.-9:30 p.m., Garden
Announcement of available cluster resources for large scale systems research for the years of 2013 and 2014. Webpage: http://www.nmc-probe.org/ Contact: firstname.lastname@example.org
The Parallel Reconfigurable Observational Environment (PRObE) is an NSF sponsored research facility housed at the New Mexico Consortium (NMC) in Los Alamos, NM. It was designed to provide large-scale, dedicated computer resources for the systems research community. All PRObE resources are dedicated for systems research and users have full (and remote) access to the system hardware and software through the Emulab (emulab.net) interface developed at the University of Utah. At the moment, three clusters are online: Marmot ~ 100 nodes, Denali ~ 100 nodes, and Kodiak ~ 1000 nodes. Two more clusters will come online in the next six months. The majority of machines are housed at the NMC in Los Alamos and a smaller subset at Carnegie Mellon University in Pittsburgh, PA. Use of the smaller clusters follows standard Emulab online proposal and sharing procedures, and are intended for porting and practice use. Use of the larger clusters is time shared, not space shared. PIs write a short proposal justifying the science and their preparation and PRObE allocates the entire resource for days to weeks. All users are required to publish and/or release experimental results gathered using PRObE resources.
Linux has a multitude of projects in the file and storage space. Come get an update from some of those directly involved, share your own projects or just ask questions.
Stop by the WiAC BoF to learn more about the USENIX Women in Advanced Computing initiative and meet and mingle with other women working in file and storage systems.
NFSv4.1 and pNFS is entering the implementation and deployment phase as it is now available in many Linux distributions. Many early adapters are already using NFSv4.1 and pNFS file, block and object layouts. This BOF will offer current information on the availability of NFSv4.1 implementations now reaching the market, and attendees will have the opportunity to provide direct feedback to major vendors on user requirements.
The Software-Defined Data Center is a unified data center platform that provides automation, flexibility, and efficiency to transform IT infrastructures. Compute, storage, networking, security, and availability services are pooled, aggregated, and delivered as software, and managed by intelligent, policy-driven software. In this Bof, we will focus on Software Defined Storage (SDS) that brings together disparate physical storage resources to be managed as a single virtual pool in a way hypervisors have transformed the physical servers. We will discuss the current landscape of SDS technologies, and outline the challenges and the opportunities for the future.