Summary

1
Summary

• Background
• Why do we need parallel processing? Moores law.
  Applications.
• Introduction in algorithms and applications
• Methodology to develop efficient parallel
  (distributed-memory) algorithms 
• Understand various forms of overhead
  (communication, load imbalance, search overhead,
  synchronization)
• Understand various distributions (blockwise,
  cyclic)
• Understand various load balancing strategies
  (static, dynamic master/worker model)
• Understand correctness problems (e.g. message
  ordering)

2
Summary

• Parallel machines and architectures
• Processor organizations, topologies, criteria
• Types of parallel machines
• arrays/vectors, shared-memory, distributed memory
• Routing
• Flynns taxonomy
• What are cluster computers?
• What networks do real machines (like the Blue
  Gene) use?
• Speedup, efficiency ( their implications),
  Amdahls law

3
Summary

• Programming methods, languages, and environments
• Different forms of message passing
• naming, explicit/implicit receive,
  synchronous/asynchronous sending
• Select statement
• SR primitives (not syntax)
• MPI message passing primitives, collective
  communication
•  
• Java parallel programming model and primitives
• HPF problems with automatic parallelization
  division of work between programmer and HPF
  compiler alignment/distribution primitives
  performance implications

4
Summary

• Applications
• N-body problems load balancing and communication
  (locality) optimizations, costzones, performance
  comparison
• Search algorithm (TDS) use asynchronous
  communication clever (transposition-driven)
  scheduling
• Grid computing
• What are grids? Why do we need them?
• Application-level tools for grid programming and
  execution
• Why is parallel computing on grids feasible?
• Ibis goals, applications, design, programming
  vs. deployment, communication primitives,
  divideconquer parallelism, performance behavior
  on a grid

5
Summary

• Multimedia content analysis on Grids
• What is Multimedia Content Analysis (MMCA) ?
  Imaging applications, feature vectors, low-level
  patterns
• Why parallel computing in MMCA - and how? Need
  for speed and transparency
• Software Platform Parallel-Horus. Task/data
  parallelism, Separable Recursive Filtering, Lazy
  Parallelization
• Example Parallel Image Processing on Clusters
• Grids and their specific problems. Promise and
  problems of the grid
• Towards a Software Platform for MMCA on Grids.
  Problems with wide-area servers
• Large-scale MMCA applications on Grids (TRECVID,
  object recognition)

6
Summary

• Multi/many-cores processors
• Variety of platforms (core types counts, memory
  architecture sizes, parallelism layers types,
  scheduling)
• Performance evaluation (arithmetic intensity,
  Roofline model)
• Programming models for multi/many-cores
• Deal with performance, portability,
  programmability
• The correlator application
• Programming models for Cell, GPUs, Generic models