Call for Position Statements
in PDF Format

2007 Linux Storage & Filesystem Workshop
February 12–13, 2007
San Jose, CA

Co-located with the 5th USENIX Conference on File and Storage Technologies (FAST '07), which will take place February 13–16, 2007

Important Dates
Submissions due: November 24, 2006
Notification of acceptance: December 15, 2006

Workshop Organizers

Program Chair
Ric Wheeler, EMC

Program Committee
Jens Axboe, Oracle
James Bottomley, SteelEye
Valerie Henson, Intel
Andrew Morton, Google
Trond Myklebust, Network Appliance
Brian Pawlowski, Network Appliance
Theodore Ts'o, IBM

Overview
The Linux Storage and Filesystem Workshop is a small, tightly focused, by-invitation workshop. It is intended to bring together developers and researchers interested in implementing improvements in the Linux filesystem and storage subsystems that can find their way into the mainline kernel and into Linux distributions in the 2–3-year timeframe. The workshop will be two days, the second day overlapping with FAST '07 tutorials. The workshop will be separated into storage and filesystem tracks, with a combined plenary session.

Topics and Submissions
Researchers and developers who are interested in attending should submit a 2–3-paragraph position statement that describes the topic or topics they would like to discuss during the workshop, and whether such a topic would suit the filesystem track, the storage track, or the plenary session. Examples of topics of interest include:

• New trends in storage technologies likely to impact Linux in the next 3–5 years
• More realistic methods of measuring filesystem and storage performance
• Proposed improvements to Linux filesystems, including, in particular:
  • Handling of storage errors
  • Filesystem repair techniques
  • Scaling to very large (terabyte) filesystems
• Progress reports on implementation of features discussed at the Portland Filesystem Summit
• Changes to the interface between the operating system and storage devices
• Proposed improvements to existing Linux storage subsystems, particularly with an emphasis on:
  • Refactoring common code out of storage subsystems and into the block layer
  • Better robustness and error recovery
  • Barrier implementations in the face of TCQ
  • Making use of storage capabilities (such as block guard or non–power-of-2 block sizes) for novel filesystem and application features
• Progress reports on implementation of features discussed at the Vancouver Storage Summit
• Userspace tools for managing storage systems (including better presentation to the user via sysfs)
• Storage futures, including:
  • New transports
  • Changes to existing standards for new storage features
  • SAS/SATA convergence
  • Do we yet have a use for Object-Based Storage Devices (OSD)?