High Performance Computing

123456789101112131415161718
Across
  1. 5. A parallel computing performance metric where the speedup of a system is measured when the workload is increased, but the problem size remains constant.
  2. 7. The process of integrating smaller functions directly into the body of their callers to reduce function call overhead.
  3. 8. Variability in a system's performance due to operating system activities, which can affect the timing and efficiency of parallel programs.
  4. 12. A high-speed storage area within the CPU used to hold small amounts of data or instructions temporarily.
  5. 13. The primary component of a computer that performs most of the processing inside a computer.
  6. 14. The architecture where instructions to be executed by the computer are stored in memory.
  7. 15. A standardized and portable message-passing system designed to function on a wide variety of parallel computers.
  8. 16. A common measure of a computer's processing speed, often represented by the number of instructions a CPU can execute per second.
  9. 17. The time taken by a system to respond to a given input or task.Throughput, clue:The amount of work a computer system can handle in a given period, often measured in transactions per second.
  10. 18. A type of computer architecture where multiple processors share a common memory space and can communicate directly.
Down
  1. 1. The practice of specifying how data should be treated in a parallel region to ensure proper execution and avoid conflicts.
  2. 2. A situation in parallel computing where different processors complete their tasks at different times, leading to inefficiencies.
  3. 3. The phenomenon where multiple processors unnecessarily contend for the same cache line, even though they are accessing different data.
  4. 4. The allocation and deallocation of memory during program execution, as opposed to static memory allocation.
  5. 6. A popular API used for shared memory parallel programming in C, C++, and Fortran, which simplifies writing multi-threaded applications.
  6. 9. A benchmarking tool used to evaluate the performance of a computer's CPU by running a variety of tests and algorithms.
  7. 10. A benchmark used to measure a system's floating-point computing power, often used in high-performance computing.
  8. 11. A temporary storage area that holds data and instructions currently being used by the processor.
  9. 12. The concept ensuring that all processors in a shared-memory system have a consistent view of memory, preventing conflicts.