Why Comparing System-Level MPSoC Mapping Approaches is Difficult: A Case Study
Authors:
- Andres GoensAGAndres Goens
- Robert KhasanovRKRobert Khasanov
- Jeronimo CastrillonJCJeronimo Castrillon
- Simon PolstraSPSimon Polstra
- Andy PimentelAPAndy Pimentel
- 1 September 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 281-288
Abstract
Software abstractions are crucial to effectively program heterogeneous Multi-Processor Systems on Chip (MPSoCs). Prime examples of such abstractions are Kahn Process Networks (KPNs) and execution traces. When modeling computation as a KPN, one of the key challenges is to obtain a good mapping, i.e., an assignment of logical computation and communication to physical resources. In this paper we compare two system-level frameworks for solving the mapping problem: Sesame and MAPS. These frameworks, while superficially similar, embody different approaches. Sesame, motivated by modeling and design-space exploration, uses evolutionary algorithms for mapping. MAPS, being a compiler framework, uses simple and fast heuristics instead. In this work we highlight the value of common abstractions, such as KPNs and traces, as a vehicle to enable comparisons between large independent frameworks. These types of comparisons are fundamental for advancing research in the area. At the same time, we illustrate how the lack of formalized models at the hardware level are an obstacle to achieving fair comparisons. Additionally, using a set of applications from the embedded systems domain, we observe that genetic algorithms tend to outperform heuristics by a factor between 1× and 5×, with notable exceptions. This performance comes at the cost of a longer computation time, between 0 and 2 orders of magnitude in our experiments.Keywords
This publication has 18 references indexed in Scilit:
- A compiler infrastructure for embedded heterogeneous MPSoCsParallel Computing, 2014
- CoEx: A novel profiling-based algorithm/architecture co-exploration for ASIP design2013
- Mapping on multi/many-core systems2013
- MAPS: Mapping Concurrent Dataflow Applications to Heterogeneous MPSoCsIEEE Transactions on Industrial Informatics, 2011
- Rapid Design Space Exploration of Application Specific Heterogeneous Pipelined Multiprocessor SystemsIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2010
- Mapping Applications to Tiled Multiprocessor Embedded SystemsSeventh International Conference on Application of Concurrency To System Design (ACSD 2007), 2007
- A Framework for System-Level Modeling and Simulation of Embedded Systems ArchitecturesEURASIP Journal on Embedded Systems, 2007
- Multiobjective optimization and evolutionary algorithms for the application mapping problem in multiprocessor system-on-chip designIEEE Transactions on Evolutionary Computation, 2006
- A Comparison of Eleven Static Heuristics for Mapping a Class of Independent Tasks onto Heterogeneous Distributed Computing SystemsJournal of Parallel and Distributed Computing, 2001
- A framework for comparing models of computationIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 1998