UCSD / BIRN Coordinating Center NAMIC Group

1
UCSD / BIRN Coordinating CenterNAMIC Group

• Site PI Mark H. Ellisman
• Site Project Lead Jeffrey S. Grethe

2
Who are we people

• Faculty
• Mark Ellisman
• Jeffrey Grethe
• Grid Application Development
• Integration of Slicer3 Execution model with Grid
  infrastructure
• Marco Ruiz
• Algorithm Development
• N-Channel Automated Colocalization Analysis

• Portal Development
• Integration of Slicer / NA-MIC Toolkit components
  w/ User Portal
• Ramil Manansala
• Application Integration
• Integration of NA-MIC Toolkit with BIRN
  environment
• Vicky Rowley
• Systems Support
• Support for NA-MIC components of BIRN grid
• Bao Nguyen

3
What do we contribute - Collaboration

• Provide collaborative portal environment

• Enable researchers to store and interact with
  large collections of federated data within a
  distributed data grid

4
What do we contribute - Algorithm Development

• N-Channel Automated Colocalization Analysis
• Quantitative tools to assess the co-distributions
  of molecular markers within individual cells and
  immuloabeled tissue slices.
• Colocalization and spatial distribution analysis
  module within Slicer3

5
Incorporating Microscopy Processing and Analysis
Into 3-D Slicer
Slice view
Surface view

• File readers to deal with variety of image file
  formats generated at NCMIR
• Developing Slicer interface for existing tools
  used by NCMIR researchers (eg. IMOD, TxBR, etc)
• Create a single, user-friendly application to
  unify image processing, visualization, analysis
  and annotation

6
What do we contribute - Grid Computation

• Provide researchers with customizable access to a
  computing environment that takes advantage of
  Grid-based infrastructures

• Heterogeneous collection of arge scale compute
  resources are available to NA-MIC researchers and
  users of the NA-MIC kit (e.g. BIRN, TeraGrid,
  local clusters)
• Provide interface module (Grid Wizard Interface)
  for the NA-MIC kit to enable distributed
  execution of algorithms from Slicer and modules
  such as BatchMake

7
A Researchers Grid

• A researchers high performance computing
  resource is usually a collection of resources
  (i.e. clusters)
• To use this collection of resources, the Client
  is responsible to provide the logic to integrate
  the the disparate clusters within the logic of
  their specific applications
• Most of the times this effort is non-trivial,
  non-reusable, non-extensible, lacks robustness
  and is far from giving the end user all the
  desired functionality

8
Grid Wizard Enterprise enabled Cluster

• Tight integration with cluster internals and
  transparent access to them.
• Granular level of execution control submission,
  pause, resume, abort.
• Real time monitoring and alerting capabilities.
• Granular reporting of historic, diagnostics and
  statistics data.
• Transparent environment translation for requests.
• Programmatic control with rich and simple API.

9
Tool Integration - Generic Slicer3
10
Immediate Goals
June 2008
February 2008
April 2008
Current
On-Going BatchMake Integration
Pre-release. Alpha version of complete
infrastructure including automatic deployment
Grid Interface prototype
Pre-release (Beta version) - Full Slicer 3
Integration
Production Release - Stable version with drivers
for most popular cluster components (including
full transparent SRB support), full monitoring,
alert capabilities and API bundles.