Processing Multiple Buffers in Parallel to Increase Performance
This paper describes how processing multiple independent data buffers in parallel can dramatically improve performance, even without SIMD. A scheduler allows this approach to be used even when the size of each individual buffer varies. The net result is 2X-3X improvement in performance over the best-known single-buffer methods on Intel processors.
Please disable any pop-up blockers for proper viewing of this Whitepaper.