Storage on the Lunatic Fringe: Beyond Peta-Scale Storage Systems
           
Event Type Start Time End Time Rm # Chair  

 

Workshop 10:30AM 5:00PM 42-43 Thomas M. Ruwart (University of Minnesota)
 
Title:
(General Info)
  Speakers/Presenter:
Organizer: Thomas M. Ruwart (University of Minnesota Digital Technology Center)

 

Workshop 10:30AM 12:00PM 42-43 Thomas M. Ruwart (University of Minnesota)
 
Title:
Current and Future HEC Application Requirements for Storage Systems
  Speakers/Presenter:
Session Chair: Steve Louis (LLNL), Panelists: Bill Boas (LLNL), Keith Gray (BP), Richard Mount (SLAC), Dan Reed (NCSA), Curt Tilmes (NASA)

 

Workshop 1:30PM 3:00PM 42-43 Thomas M. Ruwart (University of Minnesota)
 
Title:
Advanced Storage Concepts, Architectures, Technologies
  Speakers/Presenter:
Session Chair: Garth Gibson (Panasas,CMU), Panelists: Gabriel Broner (SGI), Roger Haskin (IBM), Jeff Hornung (Spinnaker Networks), David Pease (IBM), Tim Reddin (HP), Brent Welch (Panasas)

 

Workshop 3:30PM 5:00PM 42-43 Thomas M. Ruwart (University of Minnesota)
 
Title:
Roadmaps, Research Thrusts, and More Visibility For Storage Issues
  Speakers/Presenter:
Chair: Thomas M. Ruwart (University of Minnesota Digital Technology Center), Panelists: David Du (University of Minnesota), Paul Frank (INSIC), Greg Ganger (CMU), Ethan Miller (UC Santa Cruz), Giora Tarnopolsky (INSIC), plus one TBA
             

 

     
  Session: Storage on the Lunatic Fringe: Beyond Peta-Scale Storage Systems
  Title: (General Info)
  Chair: Thomas M. Ruwart (University of Minnesota)
  Time: Wednesday, November 19, 10:30AM - 5:00PM
  Rm #: 42-43
  Speaker(s)/Author(s):  
  Organizer: Thomas M. Ruwart (University of Minnesota Digital Technology Center)
   
  Description:
  This workshop will help paint a picture of the current and future application requirements that significantly push the envelope of storage systems and the concepts, architectures, and technologies being developed to meet these requirements. This includes describing current work on Object-based Storage Devices (OSD) and associated file systems and alternative approaches. This leads into a discussion of adding more intelligence into the storage devices making them aware of the data objects they store. Each presentation is expected to address one of the following questions:



How to achieve beyond 1PB/sec bandwidth from a single source (i.e. a file)

How to achieve a trillion storage accesses per second

How to manage beyond a trillion objects (i.e. files)

How to find any one object withing a trillion objects

How to find things insude any number of objects within a trillion objects

Security of a trillion objects

Where existing concepts, architectures, and technologies break down

How to get storage devices to work "for" you and not "against" you

Workshop panel member presentation abstracts, biographies, and position papers may be viewed at http://www.dtc.umn.edu/programs/DISC_slfs.html as they become available.
  Link: --
   

 

     
  Session: Storage on the Lunatic Fringe: Beyond Peta-Scale Storage Systems
  Title: Current and Future HEC Application Requirements for Storage Systems
  Chair: Thomas M. Ruwart (University of Minnesota)
  Time: Wednesday, November 19, 10:30AM - 12:00PM
  Rm #: 42-43
  Speaker(s)/Author(s):  
  Session Chair: Steve Louis (LLNL), Panelists: Bill Boas (LLNL), Keith Gray (BP), Richard Mount (SLAC), Dan Reed (NCSA), Curt Tilmes (NASA)
   
  Description:
  This first session of three will describe the requirements for peta-scale storage systems from several leading-edge data-intensive users in government and industry. Panel members will describe their existing data requirements, current approaches, expected growth curves for requirements over the next 5-10 years, and what they perceive as the most significant storage and data challenges.







Questions the panelists in this session should address include:



-What are the computing, storage and I/O characteristics of your application area(s)?



-What are your storage and I/O requirements in the next 2-4 years? The next 5-10 years?



-What are your critical data showstopper issues that tend to keep you awake at night?



-How would your work change (new methods, breakthroughs) with revolutionary improvement (one or two orders of magnitude) in storage technologies?



-Would you spend more to buy multiple low-performing technologies, or would you rather invest in innovative, higher-end technology to solve your problems?



-What can you say to potential technology vendors to convince them you would buy an innovative solution to if one became available?



-Do you think visibility of lunatic fringe storage issues is high enough within government, industry, and academia? If not, what should be done to raise it?
  Link: --
   

 

     
  Session: Storage on the Lunatic Fringe: Beyond Peta-Scale Storage Systems
  Title: Advanced Storage Concepts, Architectures, Technologies
  Chair: Thomas M. Ruwart (University of Minnesota)
  Time: Wednesday, November 19, 1:30PM - 3:00PM
  Rm #: 42-43
  Speaker(s)/Author(s):  
  Session Chair: Garth Gibson (Panasas,CMU), Panelists: Gabriel Broner (SGI), Roger Haskin (IBM), Jeff Hornung (Spinnaker Networks), David Pease (IBM), Tim Reddin (HP), Brent Welch (Panasas)
   
  Description:
  This second workshop session will describe current work by several commercial vendors on advanced storage concepts (the big picture), architectures (how things are put together to form systems), and technologies (pieces used to build the systems) for current or soon-to-be-available products that can directly address the above stated requirements.
  Link: --
   

 

     
  Session: Storage on the Lunatic Fringe: Beyond Peta-Scale Storage Systems
  Title: Roadmaps, Research Thrusts, and More Visibility For Storage Issues
  Chair: Thomas M. Ruwart (University of Minnesota)
  Time: Wednesday, November 19, 3:30PM - 5:00PM
  Rm #: 42-43
  Speaker(s)/Author(s):  
  Chair: Thomas M. Ruwart (University of Minnesota Digital Technology Center), Panelists: David Du (University of Minnesota), Paul Frank (INSIC), Greg Ganger (CMU), Ethan Miller (UC Santa Cruz), Giora Tarnopolsky (INSIC), plus one TBA
   
  Description:
  This final workshop session will explore technology roadmaps and advanced research ideas from industry and academia that will be necessary to address longer-term HEC storage system requirements. Panelists will describe how these future technologies are likely to evolve relative to one another and what will be the most significant problems to overcome (technical, monetary, and political). Finally, a group discussion will focus on a call-to-action in terms of next steps in evolving the research into development projects. Finally, a group discussion will focus on a call-to-action in terms of next steps in evolving the research into development projects.



Questions the panelists in this session should address include:

-What are the top 3 storage problems that need to be researched? Do these problems relate to: (a) Capacity, (b) Latency (time to data), (c) Bandwidth (bytes per second), (d) Reliability/Availability/Servicability (e)Management (device, network, date, information, knowledge) (f) Security, (g) Other?

-What are the dynamics (e.g., changing markets, changing needs) of the above problems?

-What are the barriers to progress on addressing these problems? (a) Standards (b) Market adoption (c) Legacy support (d) Visibility (e) Other?

-Where are some of the more promising/interesting research efforts that you are aware of (nanotubes, MEMS, holographic, etc.)? (a) Short term research (1-3 years), (b) Long term research (4-10) years

-How do you see our common view of storage changing in the next 10 years? (what will you do with a 50TByte IPOD?)
  Link: --