cfaed Publications

Domain-specific hybrid mapping for energy-efficient baseband processing in wireless networks

Reference

Robert Khasanov, Julian Robledo, Christian Menard, Andres Goens, Jeronimo Castrillon, "Domain-specific hybrid mapping for energy-efficient baseband processing in wireless networks" (to appear), Proceedings of the 2021 International Conference on Compilers, Architecture, and Synthesis of Embedded Systems (CASES), IEEE Press, pp. 1-20, Oct 2021.

Abstract

Advancing telecommunication standards continuously push for larger bandwidths, lower latencies, and faster but also with a high workload heterogeneity. As a consequence of the required flexibility, baseband processing has seen a trend towards software implementations in cloud Radio Access Networks (cRANs). The flexibility gained from software implementation comes at the price of impoverished energy efficiency. This paper data rates. The receiver baseband unit not only has to deal with a huge number of users expecting connectivity
addresses the trade-off between flexibility and efficiency by proposing a domain-specific hybrid mapping algorithm. Hybrid mapping is an established approach from the model-based design of embedded systems that allows us to retain flexibility while targeting heterogeneous hardware. Depending on the current workload, the runtime system selects the most energy-efficient mapping configuration without violating timing constraints. We leverage the structure of baseband processing to enable efficient mapping of 100s of tasks at the millisecond granularity, improving upon state-of-the-art hybrid approaches. We validate our approach on an Odroid XU4 and virtual platforms with application-specific accelerators. In our evaluation on different LTE workloads, our hybrid approach achieves, in a median execution, the same performance with the same hardware resources, while only consuming about a third of the dynamic energy, compared to a work-stealing software scheduler.

Bibtex

@InProceedings{khasanov_cases21,
author = {Robert Khasanov and Julian Robledo and Christian Menard and Andres Goens and Jeronimo Castrillon},
booktitle = {Proceedings of the 2021 International Conference on Compilers, Architecture, and Synthesis of Embedded Systems (CASES)},
title = {Domain-specific hybrid mapping for energy-efficient baseband processing in wireless networks},
pages = {1-20},
abstract = {Advancing telecommunication standards continuously push for larger bandwidths, lower latencies, and faster but also with a high workload heterogeneity. As a consequence of the required flexibility, baseband processing has seen a trend towards software implementations in cloud Radio Access Networks (cRANs). The flexibility gained from software implementation comes at the price of impoverished energy efficiency. This paper data rates. The receiver baseband unit not only has to deal with a huge number of users expecting connectivity
addresses the trade-off between flexibility and efficiency by proposing a domain-specific hybrid mapping algorithm. Hybrid mapping is an established approach from the model-based design of embedded systems that allows us to retain flexibility while targeting heterogeneous hardware. Depending on the current workload, the runtime system selects the most energy-efficient mapping configuration without violating timing constraints. We leverage the structure of baseband processing to enable efficient mapping of 100s of tasks at the millisecond granularity, improving upon state-of-the-art hybrid approaches. We validate our approach on an Odroid XU4 and virtual platforms with application-specific accelerators. In our evaluation on different LTE workloads, our hybrid approach achieves, in a median execution, the same performance with the same hardware resources, while only consuming about a third of the dynamic energy, compared to a work-stealing software scheduler.},
location = {Virtual conference},
month = oct,
numpages = {20},
publisher = {IEEE Press},
year = {2021},
}

Downloads

No Downloads available for this publication

Related Paths

Orchestration Path

Permalink

https://cfaed.tu-dresden.de/publications?pubId=3155


Go back to publications list