Loading...
Please wait, while we are loading the content...
Similar Documents
Cross-Platform Performance Prediction of Parallel Applications Using Partial Execution (2005)
| Content Provider | CiteSeerX |
|---|---|
| Author | Yang, Leo T. Ma, Xiaosong Mueller, Frank |
| Description | Performance prediction across platforms is increasingly important as developers can choose from a wide range of execution platforms. The main challenge remains to perform accurate predictions at a low-cost across different architectures. In this paper, we derive an affordable method approaching cross-platform performance translation based on relative performance between two platforms. We argue that relative performance can be observed without running a parallel application in full. We show that it suffices to observe very short partial executions of an application since most parallel codes are iterative and behave predictably manner after a minimal startup period. This novel prediction approach is observation-based. It does not require program modeling, code analysis, or architectural simulation. Our performance results using real platforms and production codes demonstrate that prediction derived from partial executions can yield high accuracy at a low cost. We also assess the limitations of our model and identify future research directions on observationbased performance prediction. 1. |
| File Format | |
| Language | English |
| Publisher Date | 2005-01-01 |
| Publisher Institution | In Proc. of the IEEE/ACM Supercomputing’2005: High Performance Networking and Computing Conference |
| Access Restriction | Open |
| Subject Keyword | Relative Performance Program Modeling Architectural Simulation Different Architecture Affordable Method Partial Execution Main Challenge Production Code Future Research Direction Observationbased Performance Prediction Cross-platform Performance Translation Code Analysis Performance Prediction Short Partial Execution High Accuracy Minimal Startup Period Accurate Prediction Real Platform Parallel Code Cross-platform Performance Prediction Novel Prediction Approach Performance Result Wide Range Parallel Application Parallel Application Using Partial Execution Execution Platform Low Cost |
| Content Type | Text |
| Resource Type | Article |
