Mumbai University · Final Year Information Technology · Rev-2019 'C' Scheme
High Performance Computing
ITDO7012 · Department Optional Course 3 · Semester VII · 3 credits · Internal tests 20 (average of two) + end-semester 80
End-semester paper (3 hours): Six questions of 20 marks; Q1 is compulsory and you answer four.
One of the Department Optional Course 3 options.
Module 1: Introduction to Parallel Computing
5 hoursMotivation and scope; levels of parallelism; SIMD, MIMD, SIMT, SPMD, dataflow and demand-driven models.
Examined in the 80-mark end-semester paper; the class tests cover what has been taught by then
- Levels of parallelism
- Parallel computing models: SIMD, MIMD, SIMT and SPMD
Module 2: Parallel Programming Platforms
4 hoursImplicit parallelism; the dichotomy of parallel platforms; physical organisation; communication costs.
Examined in the 80-mark end-semester paper; the class tests cover what has been taught by then
- Physical organisation of parallel platforms
- Communication costs in parallel machines
Module 3: Parallel Algorithms and Concurrency
9 hoursDecomposition; tasks and interactions; mapping for load balancing; broadcast and reduction operations.
Examined in the 80-mark end-semester paper; the class tests cover what has been taught by then
- Decomposition techniques
- Mapping for load balancing
- Parallel algorithm models
- Broadcast and reduction operations
Module 4: Performance Measures
5 hoursSpeedup, execution time, efficiency, cost and scalability; granularity; Amdahl's and Gustafson's laws.
Examined in the 80-mark end-semester paper; the class tests cover what has been taught by then
- Speedup: Amdahl's and Gustafson's laws
- Scalability and cost-optimality
Module 5: Programming Paradigms for HPC
9 hoursMessage-passing principles; send and receive; MPI; topologies; matrix–vector multiplication, graph algorithms and sample sort.
Examined in the 80-mark end-semester paper; the class tests cover what has been taught by then
- MPI programming model
- Point-to-point communication
- Communicators and virtual topologies
- Parallel matrix-vector multiplication
- Parallel graph algorithms and sample sort
Module 6: GPGPU Architecture and Programming
5 hoursOpenCL device architectures; introduction to OpenCL programming.
Examined in the 80-mark end-semester paper; the class tests cover what has been taught by then
- OpenCL
- OpenCL device architectures
Module contents are summarised from the official Mumbai University syllabus . AiCanCode.org is not affiliated with the University of Mumbai.