MARCO : a high-performance task mapping and routing co-optimization framework for point-to-point NoC-based heterogeneous computing systems

Heterogeneous computing systems (HCSs), which consist of various processing elements (PEs) that vary in their processing ability, are usually facilitated by the network-on-chip (NoC) to interconnect its components. The emerging point-to-point NoCs which support single-cycle-multi-hop transmission, r...

Full description

Saved in:
Bibliographic Details
Main Authors: Chen, Hui, Zhang, Zihao, Chen, Peng, Luo, Xiangzhong, Li, Shiqing, Liu, Weichen
Other Authors: School of Computer Science and Engineering
Format: Article
Language:English
Published: 2022
Subjects:
Online Access:https://hdl.handle.net/10356/155574
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-155574
record_format dspace
spelling sg-ntu-dr.10356-1555742022-03-21T09:25:28Z MARCO : a high-performance task mapping and routing co-optimization framework for point-to-point NoC-based heterogeneous computing systems Chen, Hui Zhang, Zihao Chen, Peng Luo, Xiangzhong Li, Shiqing Liu, Weichen School of Computer Science and Engineering Engineering::Computer science and engineering Network-on-Chip Mapping Heterogeneous computing systems (HCSs), which consist of various processing elements (PEs) that vary in their processing ability, are usually facilitated by the network-on-chip (NoC) to interconnect its components. The emerging point-to-point NoCs which support single-cycle-multi-hop transmission, reduce or eliminate the latency dependence on distance, addressing the scalability concern raised by high latency for long-distance transmission and enlarging the design space of the routing algorithm to search the non-shortest paths. For such point-to-point NoC-based HCSs, resource management strategies which are managed by compilers, scheduler, or controllers, e.g., mapping and routing, are complicated for the following reasons: (i) Due to the heterogeneity, mapping and routing need to optimize computation and communication concurrently (for homogeneous computing systems, only communication). (ii) Conducting mapping and routing consecutively cannot minimize the schedule length in most cases since the PEs with high processing ability may locate in the crowded area and suffer from high resource contention overhead. (iii) Since changing the mapping selection of one task will reconstruct the whole routing design space, the exploration of mapping and routing design space is challenging. Therefore, in this work, we propose MARCO, the mapping and routing co-optimization framework, to decrease the schedule length of applications on point-to-point NoC-based HCSs. Specifically, we revise the tabu search to explore the design space and evaluate the quality of mapping and routing. The advanced reinforcement learning (RL)algorithm, i.e., advantage actor-critic, is adopted to efficiently compute paths. We perform extensive experiments on various real applications, which demonstrates that the MARCO achieves a remarkable performance improvement in terms of schedule length (+44.94% ∼+50.18%) when compared with the state-of-the-art mapping and routing co-optimization algorithm for homogeneous computing systems. We also compare MARCO with different combinations of state-of-the-art mapping and routing approaches. Ministry of Education (MOE) Nanyang Technological University This work is partially supported by the Ministry of Education, Singapore, under its Academic Research Fund Tier 2 (MoE2019-T2-1-071) and Tier 1 (MoE2019-T1-001-072), and Nanyang Technological University, Singapore, under its NAP (M4082282) and SUG (M4082087). 2022-03-08T02:17:16Z 2022-03-08T02:17:16Z 2021 Journal Article Chen, H., Zhang, Z., Chen, P., Luo, X., Li, S. & Liu, W. (2021). MARCO : a high-performance task mapping and routing co-optimization framework for point-to-point NoC-based heterogeneous computing systems. ACM Transactions On Embedded Computing Systems, 20(5s), 54-. https://dx.doi.org/10.1145/3476985 1539-9087 https://hdl.handle.net/10356/155574 10.1145/3476985 2-s2.0-85115852011 5s 20 54 en MoE2019-T2-1-071 MoE2019-T1-1-072 M4082282 M4082087 ACM Transactions on Embedded Computing Systems 10.21979/N9/TGQNYS © 2021 Association for Computing Machinery. All rights reserved. This paper was published in ACM Transactions on Embedded Computing Systems and is made available with permission of Association for Computing Machinery. application/pdf
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Computer science and engineering
Network-on-Chip
Mapping
spellingShingle Engineering::Computer science and engineering
Network-on-Chip
Mapping
Chen, Hui
Zhang, Zihao
Chen, Peng
Luo, Xiangzhong
Li, Shiqing
Liu, Weichen
MARCO : a high-performance task mapping and routing co-optimization framework for point-to-point NoC-based heterogeneous computing systems
description Heterogeneous computing systems (HCSs), which consist of various processing elements (PEs) that vary in their processing ability, are usually facilitated by the network-on-chip (NoC) to interconnect its components. The emerging point-to-point NoCs which support single-cycle-multi-hop transmission, reduce or eliminate the latency dependence on distance, addressing the scalability concern raised by high latency for long-distance transmission and enlarging the design space of the routing algorithm to search the non-shortest paths. For such point-to-point NoC-based HCSs, resource management strategies which are managed by compilers, scheduler, or controllers, e.g., mapping and routing, are complicated for the following reasons: (i) Due to the heterogeneity, mapping and routing need to optimize computation and communication concurrently (for homogeneous computing systems, only communication). (ii) Conducting mapping and routing consecutively cannot minimize the schedule length in most cases since the PEs with high processing ability may locate in the crowded area and suffer from high resource contention overhead. (iii) Since changing the mapping selection of one task will reconstruct the whole routing design space, the exploration of mapping and routing design space is challenging. Therefore, in this work, we propose MARCO, the mapping and routing co-optimization framework, to decrease the schedule length of applications on point-to-point NoC-based HCSs. Specifically, we revise the tabu search to explore the design space and evaluate the quality of mapping and routing. The advanced reinforcement learning (RL)algorithm, i.e., advantage actor-critic, is adopted to efficiently compute paths. We perform extensive experiments on various real applications, which demonstrates that the MARCO achieves a remarkable performance improvement in terms of schedule length (+44.94% ∼+50.18%) when compared with the state-of-the-art mapping and routing co-optimization algorithm for homogeneous computing systems. We also compare MARCO with different combinations of state-of-the-art mapping and routing approaches.
author2 School of Computer Science and Engineering
author_facet School of Computer Science and Engineering
Chen, Hui
Zhang, Zihao
Chen, Peng
Luo, Xiangzhong
Li, Shiqing
Liu, Weichen
format Article
author Chen, Hui
Zhang, Zihao
Chen, Peng
Luo, Xiangzhong
Li, Shiqing
Liu, Weichen
author_sort Chen, Hui
title MARCO : a high-performance task mapping and routing co-optimization framework for point-to-point NoC-based heterogeneous computing systems
title_short MARCO : a high-performance task mapping and routing co-optimization framework for point-to-point NoC-based heterogeneous computing systems
title_full MARCO : a high-performance task mapping and routing co-optimization framework for point-to-point NoC-based heterogeneous computing systems
title_fullStr MARCO : a high-performance task mapping and routing co-optimization framework for point-to-point NoC-based heterogeneous computing systems
title_full_unstemmed MARCO : a high-performance task mapping and routing co-optimization framework for point-to-point NoC-based heterogeneous computing systems
title_sort marco : a high-performance task mapping and routing co-optimization framework for point-to-point noc-based heterogeneous computing systems
publishDate 2022
url https://hdl.handle.net/10356/155574
_version_ 1728433382547783680