Parameterizing Simulated Annealing for Distributing Task Graphs on Multiprocessor SoCs

Orsila, Heikki; Kangas, Tero; Salminen, Erno; Hämäläinen, Timo D.

doi:10.1109/issoc.2006.321971

Cited by 18 publications

(23 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore the final temperature must be set as high as possible without sacrificing the greedy part in optimization. This is the rationale for (Orsila, 2006) in Section 4.3.…”

Section: Saving Optimization Effortmentioning

confidence: 89%

“…This case presents methods to derive SA parameters systematically from the problem parameters (Orsila, 2006). The annealing schedule is geometric with q = 0.95.…”

Section: Orsila Casementioning

confidence: 99%

“…As k increases p f decreases exponentially. Suitable values for k are in range [1,9] unless L is very large (hundreds of thousands or even millions of iterations). The temperature range implied by k = 1 is shown in Figure 5.…”

Section: Orsila Casementioning

confidence: 99%

See 2 more Smart Citations

Best Practices for Simulated Annealing in Multiprocessor Task Distribution Problems

Orsila¹,

Salminen²,

Hämäläinen³

2008

Simulated Annealing

Self Cite

View full text Add to dashboard Cite

“…Therefore the final temperature must be set as high as possible without sacrificing the greedy part in optimization. This is the rationale for (Orsila, 2006) in Section 4.3.…”

Section: Saving Optimization Effortmentioning

confidence: 89%

“…This case presents methods to derive SA parameters systematically from the problem parameters (Orsila, 2006). The annealing schedule is geometric with q = 0.95.…”

Section: Orsila Casementioning

confidence: 99%

See 1 more Smart Citation

Best Practices for Simulated Annealing in Multiprocessor Task Distribution Problems

Orsila¹,

Salminen²,

Hämäläinen³

2008

Simulated Annealing

Self Cite

View full text Add to dashboard Cite

“…Methods based on branch-and-bound, evolutionary algorithms and constraint formulations have also been devised [25]. Simulated annealing has recently found success in multiprocessor scheduling problems [21]. Ravindran [22] shows a toolbox of methods for solving the static scheduling problem.…”

Section: Related Workmentioning

confidence: 99%

“…However, scheduling algorithms that handle constraints such as specific architecture topologies, memory limitations and so on have to deal with a complicated and not very well understood solution space. A general method that has been successfully applied to practical and complex multiprocessor scheduling problems is Simulated Annealing [14] [21].…”

Section: Introductionmentioning

confidence: 99%

Scheduling task dependence graphs with variable task execution times onto heterogeneous multiprocessors

Satish

Ravindran

Keutzer

2008

Proceedings of the 8th ACM International Conference on Embedded Software

View full text Add to dashboard Cite

We present a statistical optimization approach for a scheduling a task dependence graph with variable task execution times onto a heterogeneous multiprocessor system. Scheduling methods in the presence of variations typically rely on worst-case timing estimates for hard real-time applications, or average-case analysis for other applications. However, a large class of soft real-time applications require only statistical guarantees on latency and throughput. We present a general statistical model that captures the probability distributions of task execution times as well as the correlations of execution times of different tasks. We use a Monte Carlo based technique to perform makespan analysis of different schedules based on this model. This approach can be used to analyze the variability present in a variety of soft real-time applications, including a H.264 video processing application.We present two scheduling algorithms based on statistical makespan analysis. The first is a heuristic based on a critical path analysis of the task dependence graph. The other is a simulated annealing algorithm using incremental timing analysis. Both algorithms take as input the required statistical guarantee, and can thus be easily re-used for different required guarantees. We show that optimization methods based on statistical analysis show a 25-30% improvement in makespan over methods based on static worst-case analysis.

show abstract

Automated memory-aware application distribution for Multi-processor System-on-Chips

Orsila

Kangas

Salminen

et al. 2007

Journal of Systems Architecture

View full text Add to dashboard Cite

Parameterizing Simulated Annealing for Distributing Task Graphs on Multiprocessor SoCs

Cited by 18 publications

References 1 publication

Best Practices for Simulated Annealing in Multiprocessor Task Distribution Problems

Best Practices for Simulated Annealing in Multiprocessor Task Distribution Problems

Scheduling task dependence graphs with variable task execution times onto heterogeneous multiprocessors

Automated memory-aware application distribution for Multi-processor System-on-Chips

Contact Info

Product

Resources

About