CSC 456 Fall 2013/1d vb: Difference between revisions
Jump to navigation
Jump to search
| Line 27: | Line 27: | ||
| 500 cabinets * 24 blades * 4 cores * 3 pipelines per chip | | 500 cabinets * 24 blades * 4 cores * 3 pipelines per chip | ||
|} | |} | ||
==Sources== | |||
<ol> | |||
<li>derp</li> | |||
</ol> | |||
Revision as of 16:16, 3 September 2013
Trends in Pipelining
Trends of Increasing Pipeline Length
Trends of Decreasing Pipeline Length
The issue with increased pipeline length is the problem of incorrect branch predictions. The longer a pipeline is, the more stages of wasted processing have been wasted when a different branch is taken. Decreasing the pipeline length has resulted in lower clock frequencies, but equal or better IPC. A smaller pipeline suffers less of a loss for every bad prediction, and the overall performance is improved.
An Example of Pipeline changes in Cray Systems
| Year | Name | Pipeline Length | Number of Pipelines |
|---|---|---|---|
| 1976 | Cray 1 | 3 | 12 |
| 2012 | Cray XK7 | 12 for scalar , 17 for vector | 500 cabinets * 24 blades * 4 cores * 3 pipelines per chip |
Sources
- derp