1 | For both parallel and sequential computers, cost is accessing remote memories with some form of "communication" |
2 | Data locality addresses in both cases |
3 | Differences are quantitative size of effect and what is done by user and what automatically |
4 | Processor |
5 | Cache |
6 | L2 Cache |
7 | L3 Cache |
8 | Main |
9 | Memory |
10 | Processor |
11 | Cache |
12 | L2 Cache |
13 | Processor |
14 | Cache |
15 | L2 Cache |
16 | Board Level Interconnection Networks |
17 | .... |
18 | .... |
19 | System Level Interconnection Network |
20 | L3 Cache |
21 | Main |
22 | Memory |
23 | L3 Cache |
24 | Main |
25 | Memory |
26 | Slow |
27 | Very Slow |