Your SlideShare is downloading. ×
  • Like
Bringing The Grid Home for Grid2008
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×

Now you can save presentations on your phone or tablet

Available for both IPhone and Android

Text the download link to your phone

Standard text messaging rates apply

Bringing The Grid Home for Grid2008

  • 345 views
Published

Presentation for Bringing the Grid Home presented to Grid 2008. In this presentation I discuss my work G-ICING

Presentation for Bringing the Grid Home presented to Grid 2008. In this presentation I discuss my work G-ICING

Published in Technology , Education
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads

Views

Total Views
345
On SlideShare
0
From Embeds
0
Number of Embeds
0

Actions

Shares
Downloads
2
Comments
0
Likes
0

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide
  • Good afternoon and Konichiwa Today I’ll be talking about G-ICING, an Installable File System for Windows that interfaces with a Grid-Backend. As this photo shows, G-ICING is intended to be a layer on top of the Grid intended to make the grid tastier ;-)
  • Explosion of information systems has led to vast arrays of data stored in widely varying formats, locations, and subject to numerous access and privacy policies. Collected and housed in standalone data stores with little or no integration between systems. Definte interest in more integrations resources. Introduce the grid, integrating information across organizational boundaries. Requires tackling a wide range of issues including finding, naming and accessing, crossing firewalls, enforcing policies, et al. Grid was supposed to solve all this (move with slide)
  • Simple and familiar - users should use a familiar paradigm. Users do not like to change the way they work, instead we should extend existing paradigms to interact with the Grid rather than vice versa COTS and legacy applications do not change. We must use something that they already use Performance, we just can’t die here Security – same shpeal Standards in order to interact with others People don’t want to be locked into one vendor. Help innovate on a grand scale since everyone can talk to each other Don’t have to develop same things more than once to take advantage of others sharing resources. Simple and familiar User Transparency Application Transparency We FORGET this criterion often Performance Users and applications don’t like to wait around all day Security Organizations need to enforce policies Normal users need it Based on Standards to allow interaction between different Grid infrastructures
  • More FOCUS Talk about how people and applications are familiar with the filesystem interface. Talk about how the IFS transparently maps the Grid into the Windows file system (DO NOT GLOSS OVER SOS-KAY).
  • Uva, Fujitsu, EPCC (Edinburgh), UNICORE
  • Focus on more Inv Call Model
  • Security – within the host G-icing relies upon Windows security to protection information between users of the same machine. Host-grid secuirty is complex. G-icing acts as a proxy for users. GIS prompts user for credentials and user is given a choice of using one of their certificates. GIS gets a signed delegated credential. Uses WS-Security* family of specifications and profiles.
  • We knew would be slow. Overhead involved with WS is enormous. Multiple XML serialization and deserialization ops with signing and encryption. A factor slower (10x)‏
  • 3 – double clicking on an application etc

Transcript

  • 1. Bringing the Grid Home Chris Sosa, Andrew Grimshaw University of Virginia October 1, 2008
  • 2. Introduction
  • 3. The Big Problem
    • Grid has little uptake
    Hard to Use Inflexible Security Doesn’t play well with others
  • 4. Solution Criteria
  • 5. Bring in G-ICING
    • Users and applications are familiar with the filesystem
    • Grid-backend will appear as a network drive to users
  • 6. Talking to a Grid-Backend
    • Resource Naming Service (RNS)
      • Basic directory services
      • Entries may be added, deleted and listed
    • ByteIO
      • Provides POSIX-like interface to resources
      • Resources can be anything (database, a file, a host)
    • Who else uses it?
      • ByteIO interoperability fiesta – UVa, Fujitsu, EPCC, UNICORE
      • CERN has a working RNS with LFC and is working on a ByteIO implementation
  • 7. User to Kernel Communication
    • DESIGN ALTERNATIVES
    • Local Procedure Calls
      • Deprecated by MS
      • Calls are synchronous
    • Inverted Call Model
      • Takes advantage of the I/O mechanisms in WinNT
      • User level program makes special I/O Request: “Hello, I’m waiting for an operation” and Kernel mode stores it
      • Kernel forwards actual I/O requests, to the user mode by responding to the above I/O with the forwarded I/O call
  • 8. IFS Development in Windows
  • 9. G-ICING Design
  • 10. Kernel Management Service (KMS)
    • Installable File System Driver
      • Network Redirector
      • Kernel driver that interacts with other Kernel components
    • Communicates to User-mode UFS with Inverted Call Model
  • 11. User Forwarding Service (UFS)
    • Uses JNI to communicate and forwards requests to GIS
    • Prompts user for credentials and obtains a delegated credential for use
    Flexible Security through Delegation
  • 12. Grid Interface Service (GIS)
    • Converts FS requests into ByteIO/RNS calls
    • Caches meta-information from ByteIO/RNS
    • Only interface to Grid-backend
    Meta-data Caching
  • 13. Prototype in Action
  • 14. Prototype in Action (Continued)
  • 15. Evaluation
    • Usability (compared to the alternatives)
        • Shell Extension (not app transparent)
        • Posix-like libraries (neither user or app transparent)
        • Special libraries – application transparent only if recompiled (can't with COTS or Legacy)
    • Security
      • Depends on the Grid-backend but prototype is compatible with a flexible delegation model
    • Standards – RNS, ByteIO
    • Performance – Coming up …
  • 16. Performance Evaluation Setup
    • Client
      • Single-core 2.34 GHz desktop machine with 1GB memory running WinXP
      • 100 Mbps connection
    • Grid-Backend
      • Genesis II running on seven 8-core Xeon processors running at 2.33 GHz with 16 GB memory
      • 1 Gbps connection
  • 17. Performance Evaluation
  • 18. Future Work
    • Improve performance
      • On-disk caching of data of both ByteIO and RNS data
      • Research in location-aware caching
    • Implementation and Comparison to different FS mechanisms in other OSs
    • Stretching RNS / ByteIO interfaces to perform more Grid services
  • 19. Conclusions
    • Grids are still too difficult to use for both users and applications
    • Can extend the filesystem paradigm in Windows to bring the Grid to users
  • 20. Related Work
    • LUFS and FUSE are filesystem in user-space technologies for UNIX / Mac
      • Lack support for Windows
      • Tied to UNIX security semantics
    • Open AFS creates a modified Samba server but stuck to Samba/CIFS security model
    • Gfarm uses FUSE + syscall hook library
      • Same problems with just FUSE
      • Overly complex for Windows, requires set up of a separate Linux box to forward messages through
    • Glite provides POSIX-like interface that is neither user or application transparent
  • 21. Questions
    • ?