Predicting the Performance Impact of Increasing Memory Bandwidth for Scientific Workflows

• Nelson Mimura Gonzalez IBM T. J. Watson Research Center
• Jose Brunheroto IBM T. J. Watson Research Center
• Fausto Artico IBM T. J. Watson Research Center
• Yoonho Park IBM T. J. Watson Research Center
• Tereza Carvalho USP
• Charles Christian Miers UDESC
• Mauricio Aronne Pillon UDESC
• Guilherme Piegas Koslovski UDESC

Resumo

The disparity between the bandwidth provided by modern processors and by the main memory led to the issue known as memory wall, in which application performance becomes completely bound by memory speed. Newer technologies are trying to increase memory bandwidth to address this issue, but the fact is that the effects of increasing bandwidth to application performance still lack exploration. This paper investigates these effects for scientific workflows focusing on the definition of a performance model and on the execution of experiments to validate the rationale for the model. The main contribution is based on two observations: memory bound applications benefit more from an increase to memory bandwidth, and the effects of improving bandwidth for a particular application gradually diminish as bandwidth is increased.

Palavras-chave: Bandwidth, Random access memory, Program processors, Measurement, Memory management, Predictive models, System performance