Trebuchet: exploring TLP with dataflow virtualisation

Parallel programming has become mandatory to fully exploit the potential of multi-core CPUs. The dataflow model provides a natural way to exploit parallelism. However, specifying dependences and control using fine-grained instructions in dataflow programs can be complex and present unwanted overheads. To address this issue, we have designed TALM: a coarse-grained dataflow execution model to be used on top of widespread architectures. We implemented TALM as the Trebuchet virtual machine for multi-cores. The programmer identifies code blocks that can run in parallel and connects them to form a dataflow graph, which allows one to have the benefits of parallel dataflow execution in a Von Neumann machine, with small programming effort. We parallelised a set of seven applications using our approach and compared with OpenMP implementations. Results show that Trebuchet can be competitive with state-of-the-art technology, while providing the benefits of dataflow execution.

[1]  Lizy Kurian John,et al.  Scaling to the end of silicon with EDGE architectures , 2004, Computer.

[2]  Steven Swanson,et al.  Modeling instruction placement on a spatial architecture , 2006, SPAA '06.

[3]  Jian Li,et al.  Power-Performance Implications of Thread-level Parallelism on Chip Multiprocessors , 2005, IEEE International Symposium on Performance Analysis of Systems and Software, 2005. ISPASS 2005..

[4]  Forum Mpi.  MPI: A Message-Passing Interface , 1994 .

[5]  Gurindar S. Sohi,et al.  Program Demultiplexing: Data-flow based Speculative Parallelization of Methods in Sequential Programs , 2006, 33rd International Symposium on Computer Architecture (ISCA'06).

[6]  L. Dagum,et al.  OpenMP: an industry standard API for shared-memory programming , 1998 .

[7]  Leslie Lamport,et al.  Distributed snapshots: determining global states of distributed systems , 1985, TOCS.

[8]  Kunle Olukotun,et al.  The case for a single-chip multiprocessor , 1996, ASPLOS VII.

[9]  Vítor Santos Costa,et al.  Transactional WaveCache: Towards Speculative and Out-of-Order DataFlow Execution of Memory Operations , 2008, 2008 20th International Symposium on Computer Architecture and High Performance Computing.

[10]  Krishna M. Kavi,et al.  Scheduled Dataflow: Execution Paradigm, Architecture, and Performance Evaluation , 2001, IEEE Trans. Computers.

[11]  Arturo González-Escribano,et al.  The OpenMP source code repository , 2005, 13th Euromicro Conference on Parallel, Distributed and Network-Based Processing.

[12]  Kunle Olukotun,et al.  The Future of Microprocessors , 2005, ACM Queue.

[13]  R. M. Tomasulo,et al.  An efficient algorithm for exploiting multiple arithmetic units , 1995 .

[14]  Maurice Herlihy,et al.  Transactional Memory: Architectural Support For Lock-free Data Structures , 1993, Proceedings of the 20th Annual International Symposium on Computer Architecture.