Decoupled vector architectures
Roger Espasa, Mateo Valero · Proceedings - International Symposium on High-Performance Computer Architecture/Proceedings · 2002
The purpose of this paper is to show that using decoupling techniques in a vector processor, the performance of vector programs can be greatly improved. Using a trace driven approach, we simulate a selection of the Perfect Club programs and compare their execution time on a conventional vector architecture and on a decoupled vector architecture. Decoupling provides a performance advantage of more than a factor of two for realistic memory latencies, and even with an ideal memory system with no latency, there is still a speedup of as much as 50%. A bypassing technique between the load/store queues is introduced and we show how it can give up to an extra speedup of 22% while also reducing total memory traffic by an average of 20%. An important part of this paper is devoted to study the tradeoffs involved in choosing an adequate size for the different queues of the architecture, so that the hardware cost of the queues can be minimized while still retaining most of the performance advantages of decoupling.
Cited by 64 works. doi:10.1109/hpca.1996.501193
Authors
References
- Simultaneous multithreading
- Design and evaluation of a compiler algorithm for prefetching
- Software pipelining
- The Perfect Club Benchmarks: Effective Performance Evaluation of Supercomputers
- Simultaneous multithreading: Maximizing on-chip parallelism
- Limits on interconnection network performance
- Performance tradeoffs in multithreaded processors
- A performance study of software and hardware data prefetching schemes
- The ZS-1 central processor
- Decoupled access/execute computer architectures