Abstract
The open-source hardware/software framework TaPaSCo aims to make reconfigurable computing on FPGAs more accessible to non-experts. To this end, it provides an easily usable task-based programming abstraction, and combines this with powerful tool support to automatically implement the individual hardware accelerators and integrate them into usable system-on-chips. Currently, TaPaSCo relies on the host to manage task parallelism and perform the actual task launches. However, for more expressive parallel programming patterns, such as pipelines of task farms, the round trips from the hardware accelerators back to the host for launching child tasks, especially when exploiting data-dependent execution times, quickly add up. The major contribution of this work is the addition of on-chip task scheduling and launching capabilities to TaPaSCo. This enables not only low-latency dynamic task parallelism, it also encompasses the efficient on-chip exchange of parameter values and task results between parent and child accelerator tasks. Our solution is able to handle recursive task structures and is shown to have latency reductions of over 35x compared to the prior approaches.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Canis, A., et al.: LegUp: high-level synthesis for FPGA-based processor/accelerator systems. In: Proceedings of the 19th ACM/SIGDA International Symposium on Field Programmable Gate Arrays, pp. 33–36 (2011)
Chen, T., Srinath, S., Batten, C., Suh, G.E.: An architectural framework for accelerating dynamic parallel algorithms on reconfigurable hardware. In: 2018 51st Annual IEEE/ACM International Symposium on Microarchitecture (MICRO), pp. 55–67. IEEE (2018)
Dubucq, T., Forlini, T., Dos Reis, V.L., Santos, I.: Matrix: bench - benchmarking the state-of-the-art task execution frameworks of many-task computing (2015)
Ernsting, S., Kuchen, H.: A scalable farm skeleton for hybrid parallel and distributed programming. Int. J. Parallel Program. 42(6), 968–987 (2014). https://doi.org/10.1007/s10766-013-0269-2
Heinz, C., Hofmann, J.A., Sommer, L., Koch, A.: Improving job launch rates in the TaPaSCo FPGA middleware by hardware/software-co-design. In: 2020 IEEE/ACM International Workshop on Runtime and Operating Systems for Supercomputers (ROSS), pp. 22–30 (2020). https://doi.org/10.1109/ROSS51935.2020.00008
Heinz, C., Hofmann, J., Korinth, J., Sommer, L., Weber, L., Koch, A.: The TaPaSCo open-source toolflow. J. Signal Process. Syst. (2021). https://doi.org/10.1007/s11265-021-01640-8
Mühlbach, S., Brunner, M., Roblee, C., Koch, A.: MalCoBox: designing a 10 Gb/s malware collection honeypot using reconfigurable technology. In: 2010 International Conference on Field Programmable Logic and Applications, pp. 592–595 (2010). https://doi.org/10.1109/FPL.2010.116
Prabhakar, R., et al.: Generating configurable hardware from parallel patterns. ACM SIGPLAN Notices 51(4), 651–665 (2016)
Vinçon, T., et al.: nKV in action: accelerating KV-stores on native computation storage with near-data processing. In: Proceedings of the VLDB Endowment, vol. 13 (2020)
Xilinx Inc: Performance and resource utilization for axi4-stream interconnect rtl v1.1. https://www.xilinx.com/support/documentation/ip_documentation/ru/axis-interconnect.html#virtexuplus
Acknowledgment
This research was funded by the German Federal Ministry for Education and Research (BMBF) in project 01 IS 17091 B.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Heinz, C., Koch, A. (2021). Supporting On-Chip Dynamic Parallelism for Task-Based Hardware Accelerators. In: Derrien, S., Hannig, F., Diniz, P.C., Chillet, D. (eds) Applied Reconfigurable Computing. Architectures, Tools, and Applications. ARC 2021. Lecture Notes in Computer Science(), vol 12700. Springer, Cham. https://doi.org/10.1007/978-3-030-79025-7_6
Download citation
DOI: https://doi.org/10.1007/978-3-030-79025-7_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-79024-0
Online ISBN: 978-3-030-79025-7
eBook Packages: Computer ScienceComputer Science (R0)