Check out my first novel, midnight's simulacra!
High Performance Parallel Computing: Difference between revisions
From dankwiki
No edit summary |
No edit summary |
||
| (4 intermediate revisions by the same user not shown) | |||
| Line 4: | Line 4: | ||
* ''weak scaling'' - maximizing work performed per unit time | * ''weak scaling'' - maximizing work performed per unit time | ||
* ''strong scaling'' - minimizing time-to-solution | * ''strong scaling'' - minimizing time-to-solution | ||
* | ==Measures of parallel algorithms== | ||
* | * Cost: proccount * exectime | ||
* Overhead: Cost(P) - Cost(1) | |||
* Speedup: exectime(1) / exectime(p) | |||
* Efficiency: Speedup / p | |||
* Scalable if | |||
** Efficiency is O(1) as p approaches infinity (*not* 0) | |||
** Work(p) is linear in p | |||
** Fixed work per processor is O(1) as p approaches infinity (*not* 0) | |||
* Isoefficiency: How fast must our working set grow to maintain constant efficiency as processors are added? | |||
==Papers== | ==Papers== | ||
* [http://www.cs.utexas.edu/users/dburger/teaching/cs395t-s08/papers/5_hill.pdf Amdahl's Law in the Multicore Era] by MD Hill | * [http://www.cs.utexas.edu/users/dburger/teaching/cs395t-s08/papers/5_hill.pdf Amdahl's Law in the Multicore Era] by MD Hill | ||
* [[Lock-free algorithms]] page | |||
Latest revision as of 11:47, 29 January 2010
CSE 6230 -- High Performance Parallel Computing
- weak scaling - maximizing work performed per unit time
- strong scaling - minimizing time-to-solution
Measures of parallel algorithms
- Cost: proccount * exectime
- Overhead: Cost(P) - Cost(1)
- Speedup: exectime(1) / exectime(p)
- Efficiency: Speedup / p
- Scalable if
- Efficiency is O(1) as p approaches infinity (*not* 0)
- Work(p) is linear in p
- Fixed work per processor is O(1) as p approaches infinity (*not* 0)
- Isoefficiency: How fast must our working set grow to maintain constant efficiency as processors are added?
Papers
- Amdahl's Law in the Multicore Era by MD Hill
- Lock-free algorithms page
