Per-packet load-balanced, low-latency routing for clos-based data center networks

Cao, Jiaxin; Xia, Rui; Yang, Pengkun; Guo, Chuanxiong; Lu, Guohan; Yuan, Lihua; Zheng, Yixin; Wu, Haitao; Xiong, Yan; Maltz, David A.

doi:10.1145/2535372.2535375

Cited by 134 publications

(64 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The core switches act as bouncing switches to perform the routing [27], and commodity switches in aggregation and access layers were utilized. The same routing scheme is assigned based on the number of nodes with each pod of FT with k = 4.…”

Section: Resultsmentioning

confidence: 99%

“…The commodity switches act as bouncing switches, implementing digit reversal bouncing (DRB), an algorithm for load balancing proposed in [27] with adequate routing condition to control traffic path within the DCN to end host, hence complementing the ECMP in splitting traffic among multiple paths easier. Packet routing interaction between server 0 and 15 in Fig.…”

Section: Network Simulationmentioning

confidence: 99%

“…11b. However, the commodity switches that replaced the energy hungry enterprise switches in 3T at the upper layers are larger in quantity and are actively involved in end-to-end aggregation of bandwidth to host servers [9,10,26,27,39], resulting an increased energy consumption of the network module in FT (see Fig. 11c, d).…”

Section: Power Utilization In Information Technology Equipmentmentioning

confidence: 99%

See 2 more Smart Citations

On energy consumption of switch-centric data center networks

Popoola

Pranggono

2017

J Supercomput

View full text Add to dashboard Cite

Data center network (DCN) is the core of cloud computing and accounts for 40% energy spend when compared to cooling system, power distribution and conversion of the whole data center (DC) facility. It is essential to reduce the energy consumption of DCN to ensure energy-efficient (green) data center can be achieved. An analysis of DC performance and efficiency emphasizing the effect of bandwidth provisioning and throughput on energy proportionality of two most common switchcentric DCN topologies: three-tier (3T) and fat tree (FT) based on the amount of actual energy that is turned into computing power are presented. Energy consumption of switch-centric DCNs by realistic simulations is analyzed using GreenCloud simulator. Power-related metrics were derived and adapted for the information technology equipment processes within the DCN. These metrics are acknowledged as subset of the major metrics of power usage effectiveness and data center infrastructure efficiency, known to DCs. This study suggests that although in overall FT consumes more energy, it spends less energy for transmission of a single bit of information, outperforming 3T.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Network Simulationmentioning

confidence: 99%

Section: Power Utilization In Information Technology Equipmentmentioning

confidence: 99%

See 1 more Smart Citation

On energy consumption of switch-centric data center networks

Popoola

Pranggono

2017

J Supercomput

View full text Add to dashboard Cite

show abstract

“…Then, with ECMP [16] and packet-level load balancing mechanisms [7,12], it turns out that the congestion usually occurs at the edge and the core is free of persistent congestion [6,20].…”

Section: B How To Set a Proper Threshold?mentioning

confidence: 99%

PAC: Taming TCP Incast Congestion Using Proactive ACK Control

Chen

Lan

et al. 2014

2014 IEEE 22nd International Conference on Network Protocols

Self Cite

View full text Add to dashboard Cite

Abstract-TCP incast congestion which can introduce hundreds of milliseconds delay and up to 90% throughput degradation, severely affecting application performance, has been a practical issue in high-bandwidth low-latency datacenter networks. Despite continuous efforts, prior solutions have significant drawbacks. They either only support quite a limited number of senders (e.g., 40-60), which is not sufficient, or require non-trivial system modifications, which is impractical and not incrementally deployable.We present PAC, a simple yet very effective design to tame TCP incast congestion via Proactive ACK Control at the receiver. The key design principle behind PAC is that we treat ACK not only as the acknowledgement of received packets but also as the trigger for new packets. Leveraging datacenter network characteristics, PAC enforces a novel ACK control to release ACKs in such a way that the ACK-triggered in-flight data can fully utilize the bottleneck link without causing incast collapse even when faced with over a thousand senders.We implement PAC on both Windows and Linux platforms, and extensively evaluate PAC using small-scale testbed experiments and large-scale ns-2 simulations. Our results show that PAC significantly outperforms the previous representative designs such as ICTCP and DCTCP by supporting 40X (i.e., 40→1600) more senders; further, it does not introduce spurious timeout and retransmission even when the measured 99 th percentile RTT is only 3.6ms. Our implementation experiences show that PAC is readily deployable in production datacenters, while requiring minimal system modification compared to prior designs.

show abstract

“…Most importantly, ECMP cannot provide succor in some traffic scenarios, such as incast. When multiple flows converge on a single receiver and the edge switch become a bottleneck, even packet-level, load-aware routing [28] will not help in this setting, while DIBS can.…”

Section: Related Workmentioning

confidence: 99%

Dibs

Zarifis

Miao

Calder

et al. 2014

Proceedings of the Ninth European Conference on Computer Systems

View full text Add to dashboard Cite

Data centers must support a range of workloads with differing demands. Although existing approaches handle routine traffic smoothly, intense hotspots-even if ephemeralcause excessive packet loss and severely degrade performance. This loss occurs even though congestion is typically highly localized, with spare buffer capacity at nearby switches. In this paper, we argue that switches should share buffer capacity to effectively handle this spot congestion without the monetary hit of deploying large buffers at individual switches. Specifically, we present detour-induced buffer sharing (DIBS), a mechanism that achieves a near lossless network without requiring additional buffers at individual switches. Using DIBS, a congested switch detours packets randomly to neighboring switches to avoid dropping the packets. We implement DIBS in hardware, on software routers in a testbed, and in simulation, and we demonstrate that it reduces the 99th percentile of delay-sensitive query completion time by up to 85%, with very little impact on other traffic.

show abstract

Per-packet load-balanced, low-latency routing for clos-based data center networks

Cited by 134 publications

References 23 publications

On energy consumption of switch-centric data center networks

On energy consumption of switch-centric data center networks

PAC: Taming TCP Incast Congestion Using Proactive ACK Control

Dibs

Contact Info

Product

Resources

About