Speeding up exascale interconnection network simulations with the VEF3 trace framework

Cano-Cano, Javier; Andújar, Francisco J.; Alfaro, Francisco J.; Sánchez, José L.

doi:10.1016/j.jpdc.2019.06.013

Cited by 4 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Work in Reference 4 has shown that Gem5 with full‐system simulation mode provides accurate software performance compared to the physical hardware system. However, full‐system set up for each hardware system is arduous and slow as compared to native execution 5,6 . On the other hand, the SysCall‐Emulation mode of Gem5 provides hardware systems simulation at a higher speed with lower accuracy.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Performance prediction from simulation systems to physical systems using machine learning with transfer learning and scaling

Mankodi

Bhatt

Chaudhury

2021

Concurrency and Computation

View full text Add to dashboard Cite

Summary Selection from several computer systems with different hardware features resulting in different software performance is a critical problem to solve. The problem becomes even more challenging when access to computer systems with different features is difficult. We had proposed a novel solution, “cross performance prediction with scaling,” in our previous work. In the scaling model, we predicted the physical system's runtime using a machine learning model trained only on a performance dataset of simulation‐based systems applying a scaling factor to the predicted runtime. In this article, we propose another novel idea, “cross performance prediction with transfer learning,” that uses transfer learning to solve the same problem. This model predicts the target physical system's performance using a machine learning model trained on a combined performance dataset from simulation‐based systems and an accessible source physical system. We evaluate both the models using several benchmark algorithms from SD‐VBS and MiBench suites. Our scaling model results have achieved a prediction error of 10%–25% for general‐purpose systems, whereas the transfer learning model has higher errors in the range of 50%. We have also developed a method to extract the rules built during the decision tree model's training to predict the runtime.

show abstract

Section: Introductionmentioning

confidence: 99%

“…compared to native execution. 5,6 On the other hand, the SysCall-Emulation mode of Gem5 provides hardware systems simulation at a higher speed with lower accuracy. Therefore, our goal is to perform accurate performance prediction of physical systems using Gem5 simulation-based hardware models using SysCall-Emulation mode.…”

mentioning

confidence: 99%