Apache Hadoop India Summit 2011 talk "Adaptive Parallel Computing over Distributed Military Computing Infrastructures" by Rituraj Kumar
Upcoming SlideShare
Loading in...5
×
 

Apache Hadoop India Summit 2011 talk "Adaptive Parallel Computing over Distributed Military Computing Infrastructures" by Rituraj Kumar

on

  • 1,811 views

 

Statistics

Views

Total Views
1,811
Views on SlideShare
1,660
Embed Views
151

Actions

Likes
0
Downloads
40
Comments
1

1 Embed 151

http://d.hatena.ne.jp 151

Accessibility

Categories

Upload Details

Uploaded via as Microsoft PowerPoint

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
  • what was this presentation all about? announcing the title itself must take like 29min. at least i couldn't make a head or tail of it.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

Apache Hadoop India Summit 2011 talk "Adaptive Parallel Computing over Distributed Military Computing Infrastructures" by Rituraj Kumar Apache Hadoop India Summit 2011 talk "Adaptive Parallel Computing over Distributed Military Computing Infrastructures" by Rituraj Kumar Presentation Transcript

  • Adaptive parallel computing over distributed military computing infrastructures
    RiturajKumar
    Center of Artificial Intelligence and Robotics – DRDO Lab
  • 2
    Contents
    • Introduction
    • Adaptive Parallel Computing
    • Approaches for APC
    • MPI + DDS
    • Hadoop
    • Conclusions
  • 3
    Introduction
    • Net-Centric Paradigm of warfare needs highly compute intensive operations to be performed.
    • Parallel Computing is the only feasible solution that would guarantee reduced response time.
    • Tactical deployment would not amenable for –
    • Large Clusters in the field
    • Back-hauling
    • Intelligent use of the existing spare compute capacity of the computing devices within the cloud is essential.
  • 4
    Adaptive Parallel Computing
    Cloud of Heterogeneous Computing Devices
    MILCOM
  • 5
    Adaptive Parallel Computing
    Amdahls Law
    The Amdahl's law is concerned with the speedup achievable from an improvement to a computation that affects a proportion P of that computation where the improvement has a speedup of S.
    Where,
    S = Speedup
    P = Parallel fraction of Program
    N = Number of Processors
    If 95% of the program can be parallelized, the theoretical maximum speedup using parallel computing would be 20×, no matter how many processors are used.
  • 6
    Adaptive Parallel Computing
    Cost for parallel Computation = Computation Cost + Serialization Cost
  • 7
    APC Approaches
    MPI + DDS
    MPI Code:
    • Parallel implementation of
    complex mathematical models
    MPI Controller:
    • Execution of parallel MPI code
    over distributed network.
    DDS:
    • Reliable Communication over
    challenged network.
    DMI:
    • Identification of computing
    nodes in the distributed
    network.
    MPI Code
    MPI
    Controller
    Dynamic Membership Identifier
    DDS
  • 8
    APC Approaches
    MPI + DDS
    Advantage:
    • Well-known MPI API Framework.
    • DDS provides assurance of reliable communication between nodes.
    Disadvantage:
    • Needs a wrapper for converting MPI calls to DDS framework.
    • Performance degradation because of the wrapper.
  • 9
    APC Approaches
    HADOOP
    • Map/Reduce can be effectively used for parallelization
    • Used extensively in varieties of information systems.
    • Java Based
    • Designed to handle large data sets
  • 10
    APC Approaches
    Hadoop
    Advantage:
    • Highly Scalable
    • Excellent fault tolerance capabilities
    • Good hardware interoperability
    Current challenges:
    • Requirement of slight architectural changes
    • Currently not suitable for resource constraint hardware
  • 11
    Conclusion
    • Military domain requires reliable parallel computing infrastructure over disadvantaged communication network.
    • Dynamic topology poses great challenge for computing infrastructure.
    • MPI framework has few disadvantages
    • Hadoop is a promising candidate in these conditions.
  • 12
    Thank You