多核与多线程.ppt

CS267 Lecture 2 * 流水图示加上CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * * * Approaches: 1) Threading 2) Data parallel 3) Transactional 4) Declarative / functional CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * Design is a minimal extension to trad processor. SMT on low end of resource replication scale. Share single execution pipeline. CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CMP enables…CMP at other end of resource replication scale. CMPs TEND to have this organization, due to performance/area tradeoffs. External interfaces to devices and signals are shared. Duplicated, independent pipelines. Later in the evaluation we will see contention from the shared resources that is relieved by the CMP’s duplication. The resource duplication can occur due to the area efficiency of CMPs, can provide private L1s and TLBs. CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * 影响所在，数据局部性和线程亲和性的重要；并行粒度的管理CS267 Lecture 2 * 应该有个Woodcrest 的图这样就好看了，而且要解释一下测试数据为什么会这个样子；充分说明分布化、层次化（一个处理器内多核是有区别的）CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * 为什么提多线程而不是多进程：多线程共享代码和数据段方便通信，调度开销低。多进程依靠SystemV 的进程间通信机制，相对复杂。CS267 Lecture 2 * CS267 Lecture 2 * 应该都介绍一下CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * CS267 Lecture 2 * * Problems in Transactional Memory * Solutions