发明名称 Scatter-gather intelligent memory architecture for unstructured streaming data on multiprocessor systems
摘要 A scatter/gather technique optimizes unstructured streaming memory accesses, providing off-chip bandwidth efficiency by accessing only useful data at a fine granularity, and off-loading memory access overhead by supporting address calculation, data shuffling, and format conversion.
申请公布号 US8966180(B2) 申请公布日期 2015.02.24
申请号 US201313782515 申请日期 2013.03.01
申请人 Intel Corporation 发明人 Kim Daehyun;Hughes Christopher J.;Chen Yen-Kuang;Kundu Partha
分类号 G06F12/00;G11C7/10;G06F12/08 主分类号 G06F12/00
代理机构 Trop, Pruner & Hu, P.C. 代理人 Trop, Pruner & Hu, P.C.
主权项 1. A processor comprising: a core including a computation processor and a scatter/gather engine coupled to the computation processor, the scatter/gather engine to generate sub-cache line sized non-sequential data accesses to a system memory based on a data access pattern to the system memory, and to communicate sub-cache line sized data with the system memory, the data access pattern comprising a program defined access pattern, wherein the scatter/gather engine includes an access pattern generator to generate the sub-cache line sized non-sequential data accesses to the system memory based on the program defined access pattern comprising a stride-based access pattern; a cache coupled to the core, data to be transferred between the cache and the system memory using full-cache line sized transfers; and a memory controller coupled between the cache and the system memory.
地址 Santa Clara CA US