Tartan: Evaluating Spatial Computation For Whole Program Execution

  • Mahim Mishra ,
  • Timothy J. Callahan ,
  • Tiberiu Chelcea ,
  • Girish Venkataramani ,
  • Mihai Budiu

International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) |

Spatial Computing (SC) has been shown to be an energy-efficient model for implementing program kernels. In this paper we explore the feasibility of using SC for more than small kernels. To this end, we evaluate the performance and energy efficiency of entire applications on Tartan, a general-purpose architecture which integrates a reconfigurable fabric (RF) with a superscalar core. Our compiler automatically partitions and compiles an application into an instruction stream for the core and a configuration for the RF. We use a detailed simulator to capture both timing and energy numbers for all parts of the system.

Our results indicate that a hierarchical RF architecture, designed around a scalable interconnect, is instrumental in harnessing the benefits of spatial computation. The interconnect uses static configuration and routing at the lower levels and a packet-switched, dynamically-routed network at the top level. Tartan is most energyefficient when almost all of the application is mapped to the RF, indicating the need for the RF to support most general-purpose programming constructs. Our initial investigation reveals that such a system can provide, on average, an order of magnitude improvement in energy-delay compared to an aggressive superscalar core on single-threaded workloads.