DAGS: Distribution agnostic sequential Monte Carlo scheme for task execution time estimation

This paper addresses the problem of stochastic task execution time estimation agnostic to the process distributions. The proposed method is orthogonal to the application structure and underlying architecture. We build the time varying state space model of the task execution time. In the case of software pipelined tasks, to refine the estimate quality, the state-space is modeled as Multiple Input Single Output (MISO) system by taking into account the current execution time of the predecessor task. To obtain nearly Bayesian estimates, irrespective of the process distribution, the sequential Monte Carlo method is applied which form the recursive solution to reduce the overheads and comprises of time update and correction steps. We experimented on three different platforms, including multicore, using the time parallelized H.264 decoder: a control dominant computationally demanding application and AES encoder: a pure data flow application. Results show that estimates obtained by our method are superior in quality and are up to 68% better in comparison to others.

[1]  Panlop Zeephongsekul,et al.  A Method for Estimating the Execution Time of a Parallel Task on a Grid Node , 2005, EGC.

[2]  Neil J. Gordon,et al.  A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

[3]  Twan Basten,et al.  Execution-time Prediction for Dynamic Streaming Applications with Task-level Parallelism , 2007, 10th Euromicro Conference on Digital System Design Architectures, Methods and Tools (DSD 2007).

[4]  Chang-Gun Lee,et al.  Stochastic analysis of periodic real-time systems , 2002, 23rd IEEE Real-Time Systems Symposium, 2002. RTSS 2002..

[5]  Yi Ma,et al.  Efficient Transient-Fault Tolerance for Multithreaded Processors Using Dual-Thread Execution , 2006, 2006 International Conference on Computer Design.

[6]  M. Pitt,et al.  Filtering via Simulation: Auxiliary Particle Filters , 1999 .

[7]  Guillem Bernat,et al.  WCET analysis of probabilistic hard real-time systems , 2002, 23rd IEEE Real-Time Systems Symposium, 2002. RTSS 2002..

[8]  Petru Eles,et al.  Schedulability analysis of multiprocessor real-time applications with stochastic task execution times , 2002, ICCAD 2002.

[9]  Petru Eles,et al.  Schedulability Analysis of Real-Time Systems with Stochastic Task Execution Times , 2002 .

[10]  Yuval Rabani,et al.  Allocating bandwidth for bursty connections , 1997, STOC '97.

[11]  Kristine L. Bell,et al.  A Tutorial on Particle Filters for Online Nonlinear/NonGaussian Bayesian Tracking , 2007 .

[12]  Lee C. Potter,et al.  Statistical prediction of task execution times through analytic benchmarking for scheduling in a heterogeneous environment , 1999, Proceedings. Eighth Heterogeneous Computing Workshop (HCW'99).

[13]  Jakob Engblom,et al.  The worst-case execution-time problem—overview of methods and survey of tools , 2008, TECS.

[14]  Arnaud Doucet,et al.  A survey of convergence results on particle filtering methods for practitioners , 2002, IEEE Trans. Signal Process..

[15]  E. Kamen,et al.  Introduction to Optimal Estimation , 1999 .

[16]  Alan Burns,et al.  A Probabilistic Framework for Schedulability Analysis , 2003, EMSOFT.

[17]  Jeanne Ferrante,et al.  Determining asynchronous acyclic pipeline execution times , 1996, Proceedings of International Conference on Parallel Processing.