Information Technology is coming soon
We're still writing and reviewing the notes for every subject, and we'll open it when every subject is ready. Meanwhile you can practise aptitude, ask Aria about any subject, or see the branches that are ready.
Information Technology
High Performance Computing
Parallel architectures and speedup, OpenMP, MPI, CUDA and OpenCL, parallel algorithms.
Notes for High Performance Computing are being written and reviewed. Until they're up, Aria teaches every topic — ask for any part of the syllabus, with worked numericals.
Coming next in High Performance Computing
- Flynn's taxonomy· notes coming
- Shared and distributed memory systems· notes coming
- Cache coherence and NUMA· notes coming
- Interconnection networks· notes coming
- Speedup: Amdahl's and Gustafson's laws· notes coming
- OpenMP fork–join model and directives· notes coming
- Work-sharing constructs and scheduling· notes coming
- Synchronisation in OpenMP· notes coming
- Data scoping and reduction· notes coming
- OpenMP tasks· notes coming
- MPI programming model· notes coming
- Point-to-point communication· notes coming
- Non-blocking communication· notes coming
- Collective communication· notes coming
- Communicators and virtual topologies· notes coming
- GPU architecture and memory hierarchy· notes coming
- CUDA kernels, threads, blocks and grids· notes coming
- CUDA memory management· notes coming
- CUDA performance optimisation· notes coming
- OpenCL· notes coming
- Parallel search: BFS and DFS· notes coming
- Parallel sorting· notes coming
- Parallel computing for AI and ML· notes coming
- Process-processor mapping· notes coming
- Decomposition techniques· notes coming
- Mapping for load balancing· notes coming
- Parallel algorithm models· notes coming
- OpenACC· notes coming
- Sources of overhead in parallel programs· notes coming
- Scalability and cost-optimality· notes coming
- Parallel matrix multiplication and Cannon's algorithm· notes coming
- Parallel all-pairs shortest paths· notes coming
- Levels of parallelism· notes coming
- Parallel computing models: SIMD, MIMD, SIMT and SPMD· notes coming
- Physical organisation of parallel platforms· notes coming
- Communication costs in parallel machines· notes coming
- Broadcast and reduction operations· notes coming
- Parallel matrix-vector multiplication· notes coming
- Parallel graph algorithms and sample sort· notes coming
- OpenCL device architectures· notes coming
- Conditions of parallelism: data and resource dependences· notes coming
- Control-flow versus data-flow computers· notes coming
- Linear pipelines and reservation tables· notes coming
- Vector processing, VLIW and array processors· notes coming
- PRAM model and basic parallel algorithms· notes coming
- Parallel prefix computation· notes coming
- Data routing operations and greedy routing· notes coming
- Mesh and hypercube algorithms· notes coming
- Butterfly, shuffle-exchange and cube-connected cycles networks· notes coming
- Multithreading and latency hiding· notes coming
- Parallel I/O and parallel file systems· notes coming
- Mapping threads to multidimensional data· notes coming
- Thread scheduling and latency tolerance· notes coming
- Warps, memory bandwidth and resource partitioning· notes coming
- GPU acceleration for neural networks· notes coming