Contention Minimization in Emerging SMART NoC via Direct and Indirect Routes

Chen, Peng; Chen, Hui; Zhou, Jun; Li, Mengquan; Liu, Weichen; Xiao, Chen; Xie, Yiyuan; Guan, Nan

doi:10.1109/tc.2021.3111517

Cited by 3 publications

(2 citation statements)

References 40 publications

(84 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We show the schedule length comparison on 4×4, 8×8, 16×16 system in Figure 6. In this experiment, we also compare our proposed algorithms with the adaptive routing [29], denoted as "adaptive", which can balance the loads for NoC. Since "dual" and our proposed algorithms take advantage of the parallel transmission, in all cases, they show better performance than the "adaptive", except for the fmradio in 16×16 mesh size.…”

Section: A Evaluation Results For Real Applicationsmentioning

confidence: 99%

“…Adaptive routing [27], [28] needs to load balance traffic and has the out-oforder packet delivery problem which is similar to multipath transmission. The algorithm proposed in [29] computes more than one candidate route for one message and applies one in the runtime based on NoC utilization to balance the loads and avoid the contention. However, compared with LAMP, the two dimensions, i.e., the number of sub-message that can be transmitted simultaneously and the size of each sub-message, are not discussed in the traditional adaptive routing algorithms.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

LAMP: Load-Balanced Multipath Parallel Transmission in Point-to-Point NoCs

Chen

Luo

et al. 2022

IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst.

Self Cite

View full text Add to dashboard Cite

Network-on-Chip (NoC) is an emerging paradigm that is able to connect a significant amount of processing elements (PEs). However, as a distributed sub-system, NoC resources have not been exploited to the fullest. Multipath parallel transmission, which splits one message into multiple parts and sends them simultaneously, shows its efficiency in utilizing NoC resources and further reducing the transmission latency. However, this method is not fully optimized in previous works, especially for emerging point-to-point NoCs due to the following reasons: (i) only limited shortest paths are chosen; (ii) static message splitting strategy without considering NoC utilization state increases contentions;(iii) the optimization of hardware that supports multipath parallel transmission is missing, resulting in additional overheads. Thus, we propose LAMP, a software and hardware collaborated design to efficiently utilize resources and reduce latency in point-to-point NoCs through the load-balanced multipath parallel transmission. Specifically, we propose a reinforcement learning-based algorithm to decide when and how to split messages, and which path should be used according to traffic loads. Also, the temporal and spatial load-balancing algorithms are proposed so that the message size is adjusted properly to utilize NoC resources. Moreover, we revise the hardware design to support multipath parallel transmission efficiently. Extensive experiments show that our algorithm achieves a remarkable performance improvement (+18.0% ∼ +29.9%) when compared with the state-of-the-art dual-path algorithm. Our hardware design decreases power and area consumption by 23.2% and 10.3% over the dual-path hardware.

show abstract

Section: A Evaluation Results For Real Applicationsmentioning

confidence: 99%