当前位置: X-MOL 学术Int. J. Parallel. Program › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
HARE: History-Aware Adaptive Routing Algorithm for Endpoint Congestion in Networks-on-Chip
International Journal of Parallel Programming ( IF 1.5 ) Pub Date : 2018-12-12 , DOI: 10.1007/s10766-018-0614-6
Kang Jin , Cunlu Li , Dezun Dong , Binzhang Fu

Endpoint congestion is one of the most challenging issues when designing low latency and high bandwidth on-chip interconnection networks. Tree saturation and head-of-line blocking caused by the endpoint congestion seriously decrease system throughput and increases network latency, leading to overall performance degradation. Adaptive routing algorithms utilize dynamic network states to route packets around congestion areas and potentially mitigate network congestions, but still cannot deal with endpoint congestions. Existing adaptive routing algorithms mainly take the current route information into account, and rarely use the route information of past packets. In this paper, we explore the route information of past packets, and led to the following novel observations that the virtual channel (VC) allocations of prior packets can be collected as useful information, and the tree saturation can be isolated through better VC selection strategy based on the past route information. Based on this observation, a novel history-aware adaptive routing algorithm for endpoint congestion, HARE, is proposed to improve network performance. We implement HARE based on the state-of-the-art routing algorithm, Footprint, and conduct extensive simulation experiments to compare it with our algorithm. The evaluation results show that our design alleviate the impact of tree saturation consistently and achieve high throughput on both synthetic and trace-driven workloads.

中文翻译:

HARE:针对片上网络中端点拥塞的历史感知自适应路由算法

在设计低延迟和高带宽的片上互连网络时,端点拥塞是最具挑战性的问题之一。端点拥塞导致的树饱和和队头阻塞严重降低系统吞吐量并增加网络延迟,导致整体性能下降。自适应路由算法利用动态网络状态在拥塞区域周围路由数据包并可能缓解网络拥塞,但仍然无法处理端点拥塞。现有的自适应路由算法主要考虑当前路由信息,很少使用过去报文的路由信息​​。在本文中,我们探索过去数据包的路由信息​​,并导致了以下新颖的观察结果,即可以收集先前数据包的虚拟通道 (VC) 分配作为有用信息,并且可以通过基于过去路由信息的更好的 VC 选择策略来隔离树饱和度。基于这一观察,提出了一种用于端点拥塞的新型历史感知自适应路由算法 HARE,以提高网络性能。我们基于最先进的路由算法 Footprint 实现了 HARE,并进行了大量的模拟实验以将其与我们的算法进行比较。评估结果表明,我们的设计始终如一地减轻了树饱和的影响,并在合成和跟踪驱动的工作负载上实现了高吞吐量。并且可以根据过去的路由信息​​通过更好的VC选择策略来隔离树的饱和度。基于这一观察,提出了一种用于端点拥塞的新型历史感知自适应路由算法 HARE,以提高网络性能。我们基于最先进的路由算法 Footprint 实现了 HARE,并进行了大量的模拟实验以将其与我们的算法进行比较。评估结果表明,我们的设计始终如一地减轻了树饱和的影响,并在合成和跟踪驱动的工作负载上实现了高吞吐量。并且可以根据过去的路由信息​​通过更好的VC选择策略来隔离树的饱和度。基于这一观察,提出了一种用于端点拥塞的新型历史感知自适应路由算法 HARE,以提高网络性能。我们基于最先进的路由算法 Footprint 实现了 HARE,并进行了大量的模拟实验以将其与我们的算法进行比较。评估结果表明,我们的设计始终如一地减轻了树饱和的影响,并在合成和跟踪驱动的工作负载上实现了高吞吐量。并进行广泛的模拟实验以将其与我们的算法进行比较。评估结果表明,我们的设计始终如一地减轻了树饱和的影响,并在合成和跟踪驱动的工作负载上实现了高吞吐量。并进行广泛的模拟实验以将其与我们的算法进行比较。评估结果表明,我们的设计始终如一地减轻了树饱和的影响,并在合成和跟踪驱动的工作负载上实现了高吞吐量。
更新日期:2018-12-12
down
wechat
bug