Birds-of-a-Feather Sessions

Lead or attend a BoF! Meet with your peers! Present new work! Don't miss these special activities designed to maximize the value of your time at the conference. The always popular evening Birds-of-a-Feather sessions are very informal gatherings of persons interested in a particular topic.

Vendor BoFs

Want to demonstrate a new product or discuss your company's latest technologies with FAST '14 attendees? Host a Vendor BoF! These sponsored one-hour sessions give companies a chance to talk about products and proprietary technology—and they include promotional benefits. Email sponsorship@usenix.org if you're interested in sponsoring a Vendor BoF. Click here for more information about sponsorship opportunities.

Scheduling a BoF

BoFs may be scheduled in advance by contacting bofs@usenix.org with "FAST '14 BoF" in the subject line and the following information in the body of the email:

  1. BoF title
  2. Organizer name and affiliation
  3. Date and time preference
  4. Brief description of BoF (optional)

BoF Schedule

Monday, February 17, 2014

ROOM

 #
of seats		 7:00 p.m.–
8:00 p.m.		 8:00 p.m.–
9:00 p.m.		 9:00 p.m.–
10:00 p.m.		 10:00 p.m.–
11:00 p.m.

Camino Real Room

(A/V includes projector/screen)

 60

Exablox Vendor BoF:

Exablox Smashfs: Architecture of a Distributed Peer-to-Peer Scale Out
Object Store with Posix Filesystem Semantics

Tad Hunt, Exablox CTO

      
Tuesday, February 18, 2014

ROOM

 #
of seats		 7:30 p.m.–
8:30 p.m.		 8:30 p.m.–
9:30 p.m.		 9:30 p.m.–
10:30 p.m.		 10:30 p.m.–
11:30 p.m.

Camino Real Room 

(A/V includes projector/screen/1 handheld mic)

 60

Vendor BoF:

EMC Reception with the Data Protection & Availability Division

 SMB3 ('CIFS') Interoperable Implementations    

Bayshore

(No A/V)

 50 Long term storage   PRObE Update: 1000 node Systems Availability & Path forward 
 
Wednesday, February 19, 2014

ROOM

 #
of seats		 7:30 p.m.–
8:30 p.m.		 8:30 p.m.–
9:30 p.m.		 9:30 p.m.–
10:30 p.m.		 10:30 p.m.–
11:30 p.m.

Camino Real Room

(A/V includes projector/screen/ 1 handheld mic)

 60

PernixData Vendor BoF  

VMWare Vendor BoF: Software-Defined Storage State of the Union

    

Bayshore West

(No A/V)

 60 USENIX Women in Advanced Computing (WiAC) BoF SNIA Programming Model for Optimal use of Persistent Memory     

Bayshore East

(No A/V)

 60 SIRF and the SNIA Long Term Retention TWG      

BoF Descriptions

Exablox Smashfs: Architecture of a Distributed Peer-to-Peer Scale Out Object Store with Posix Filesystem Semantics
Tad Hunt, Exablox CTO
Monday, February 17, 2014, 7:00 p.m.–8:00 p.m., Camino Real Room

EMC Vendor BoF
Data Protection and Availability Division, EMC
Tuesday, February 18, 2014, 7:30 p.m. - 8:30 p.m. Camino Real Room
Join the EMC Data Protection and Availability Division for food and drink following the Poster Session.

 

Long Term Storage
David Rosenthal
Tuesday, February 18, 2014, 7:30 p.m.-8:30 p.m., Bayshore

Storing large amounts of data for a long time poses both technical and economic problems. The technical problem is that the reliability implied by, for example keeping a petabyte for a century with a 50% chance of every bit being undamaged, is not merely beyond our capability to implement but beyond our our ability to measure. Up to the 2011 floods in Thailand, the 30-year history of exponential drops in disk prices meant that if you could afford to store data for a few years you could afford to store it forever. The economic problem is that this is no longer true. The disk industry's projection is that future price drops will be much slower than before the floods, which raises the cost of storing data for the long term dramatically. 

PRObE Update: 1000 node Systems Availability & Path forward
Andree Jacobson, New Mexico Consortium
Garth Gibson, Carnegie Mellon University
Tuesday, February 18, 2014, 8:30 p.m.-9:30 p.m., Bayshore

PRObE - The Parallel Reconfigurable Observational Environment is an NSF sponsored center providing large scale testbeds dedicated to low level systems research. The center has been available for about 18 months and a major technology refresh is currently underway. We're excited to announce information about our new and upcoming systems and their availability for 2014. In particular Nome (512 nodes, 8196 total cores) currently being built, and Susitna (34 nodes, FDR Infiniband, 40GbE, SSDs, and NVidia GPUs) that is currently available. 
 
NSF's PRObE (www.nmc-probe.org) operates several clusters to support systems research at large scale. The currently largest system is Kodiak, with a 1000 nodes (two core x86, 8GB DRAM, two 1TB disks, 1GE and 8Gbps IB) donated by Los Alamos National Laboratory. Full information about all clusters is available at https://www.nmc-probe.org/wiki/Machines. Today PRObE is hosting researchers from Georgia Tech, Carnegie Mellon, Los Alamos, and other institutions. Princeton researchers have published results from Kodiak at the most recent NSDI (Wyatt Lloyd, "Stronger Semantics for Low-Latency Geo-Replicated Storage", NSDI 2013).  On PRObE staging clusters are researchers from U Central Florida, UT Austin, Georgia Tech and Carnegie Mellon. For a more complete list of publications, please see https://www.nmc-probe.org/research/publications.
PRObE resources are intended for (infrastructure) systems researchers committed to public release of their research results, typically publishing in distributed systems (eg. OSDI or SOSP), cloud computing (e.g. SOCC), supercomputing (e.g. SC or HPDC), storage (e.g. FAST), or networking (e.g. NSDI). PRObE resources are managed by Emulab (www.emulab.org) a cluster manager for allocating physical nodes that has been in use for systems research for over a decade (Brian White, "An Experimental Environment for Distributed Systems and Networks," OSDI 2002).

Users start by porting and demonstrating their code on a 100-node staging cluster such as Marmot built from the same equipment donation from Los Alamos. With demonstrated success on a staging cluster, and a compelling research goal, Kodiak can be requested and allocated, possibly exclusively, for hours to days.
NSF PRObE resources will be available for at least the next two years. All uses of PRObE resources are obligated to publish their results, either in conferences or one their web sites, and acknowledge NSF PRObE resources used in these publications. See also our PRObE introduction article in the June 2013 USENIX ;login: vol 38, no 3, 2013 (www.cs.cmu.edu/~garth/papers/07_gibson_036-039_final.pdf).

 

PernixData Vendor BoF
Mahesh Patil & Woon Jung, PernixData
Wednesday, February 19, 2014, 7:30 p.m.-8:30 p.m., Camino Real Room

Capacity and Performance. When it comes to virtualized data centers both capacity and performance is usually relegated to the backend storage. Server side flash provides new options to decouple performance from capacity and achieve scale out performance. In this BOF, we present a case for server flash in SDDC, technology implications of marrying hypervisors and server flash. We present some empirical data, challenges and discuss open issues.

PernixData is fundamentally changing how storage is designed and operated in virtual data centers. The company’s flagship product, PernixData FVP, virtualizes server side flash to enable scale-out storage performance that is independent of capacity. No changes are required to VMs, servers or primary storage, ensuring maximum performance of all virtualized applications in a seamless, scalable and cost- effective manner.

VMware Vendor BoF: Software-Defined Storage State of the Union
Richard McDougall, Chief Architect - VMware
Wednesday, February 19, 2014, 8:30 p.m.-9:30 p.m., Camino Real Room

2014 is the year of Software-Defined Storage and VMware is at the forefront of this innovative approach. In this session, we will review the major technology approaches to building a software-defined storage environment with reliable and scalable storage from commodity storage servers. These technologies lay the foundation for significantly simpler and lower cost solutions. We will also look at different approaches to building software-defined storage, as well as options for layering key data services upon this foundation -- including databases, file services, big data (Hadoop HDFS) and other layered distributed storage solutions.

SNIA Programming Model for Optimal use of Persistent Memory
Doug Voigt, HP and Paul von Behren, Intel
Wednesday, February 19, 2014, 8:30 p.m.-9:30 p.m., Bayshore West

The SNIA NVM Programming Model specification describes behavior used by software to access emerging features for traditional block NVM (SSDs) and a new programming model for persistent memory (PM). PM is NVM hardware designed to be treated by software similarly to system memory using byte-addressable load/store commands rather than block oriented read/write commands.  This BOF will begin with an overview of the specification and the next work being considered in SNIA.  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.

SIRF and the SNIA Long Term Retention TWG 
Sam Fineberg, HP and Simona Rabinovici-Cohen, IBM Research
Wednesday, February 19, 2014, 7:30 p.m.-8:30 p.m., Bayshore East

Many organizations are facing the serious challenge of economically preserving and retaining access to a wide variety of digital content for dozens of years.  Long-term digital information is vulnerable to issues that do not exist in a short-term or paper world, such as media and format obsolescence, bit-rot, and loss of metadata.  Ironically, as the world becomes more digital, we may enter a “Digital Dark Age" in which business, public, and personal assets are in ever greater danger of being lost. 
  
The Storage Networking Industry Association (SNIA) Long Term Retention Technical Working Group (LTR-TWG) is addressing the "grand technical challenges" of long term digital information retention & preservation, namely both physical ("bit") and logical preservation. A major component of the TWG's Program of Work is the creation of a logical container format, named the Self-contained Information Retention Format (SIRF), for the long-term storage of digital information.  In this BoF we will present an overview of our existing work and activities, and we will seek input and participation from those attending.