DX: Latency-Based Congestion Control for Datacenters

Lee, Changhyun; Park, Chunjong; Jang, Keon; Moon, Sang Heup; Han, Dongsu

doi:10.1109/tnet.2016.2587286

Cited by 71 publications

(45 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A recent scheme, DX [17], independently identified the benefits of using delay as congestion signal for high throughput and low latency datacenter communications. DX implements accurate latency measurements using a DPDK driver for the NIC and the congestion control algorithm is within the Linux TCP stack.…”

Section: Related Workmentioning

confidence: 99%

Timely

Mittal

Lam

Dukkipati

et al. 2015

SIGCOMM Comput. Commun. Rev.

191

View full text Add to dashboard Cite

Datacenter transports aim to deliver low latency messaging together with high throughput. We show that simple packet delay, measured as round-trip times at hosts, is an effective congestion signal without the need for switch feedback. First, we show that advances in NIC hardware have made RTT measurement possible with microsecond accuracy, and that these RTTs are sufficient to estimate switch queueing. Then we describe how TIMELY can adjust transmission rates using RTT gradients to keep packet latency low while delivering high bandwidth. We implement our design in host software running over NICs with OS-bypass capabilities. We show using experiments with up to hundreds of machines on a Clos network topology that it provides excellent performance: turning on TIMELY for OS-bypass messaging over a fabric with PFC lowers 99 percentile tail latency by 9X while maintaining near line-rate throughput. Our system also outperforms DCTCP running in an optimized kernel, reducing tail latency by 13X. To the best of our knowledge, TIMELY is the first delay-based congestion control protocol for use in the datacenter, and it achieves its results despite having an order of magnitude fewer RTT signals (due to NIC offload) than earlier delay-based schemes such as Vegas.

show abstract

Section: Related Workmentioning

confidence: 99%

Timely

Mittal

Lam

Dukkipati

et al. 2015

SIGCOMM Comput. Commun. Rev.

191

View full text Add to dashboard Cite

show abstract

“…Delay-based congestion control mechanisms had continuous attention in the past for the Internet [9,39,42] and wireless networks [31,15]. Lately, refreshed interest in those mechanisms has grown in the context of DCNs [29,32]. DX [29] and TIMELY [32] measure queuing delays at the microsecond granularity, and use the measurements to keep buffer occupancy low.…”

Section: Related Workmentioning

confidence: 99%

“…Lately, refreshed interest in those mechanisms has grown in the context of DCNs [29,32]. DX [29] and TIMELY [32] measure queuing delays at the microsecond granularity, and use the measurements to keep buffer occupancy low. These approaches are single-path protocols; they may suffer from poor loadbalancing of ECMP as DCTCP does.…”

Section: Related Workmentioning

confidence: 99%

AMP: an adaptive multipath TCP for data center networks

Kheirkhah

Lee

2019

2019 IFIP Networking Conference (IFIP Networking)

View full text Add to dashboard Cite

In recent years several multipath data transport mechanisms, such as MPTCP and XMP, have been introduced to effectively exploit the path diversity of data center networks (DCNs). However, these multipath schemes have not been widely deployed in DCNs. We argue that two key factors among others impeded their adoption: TCP incast and minimum window syndrome. First, these mechanisms are ill-suited for workloads with a many-to-one communication pattern, commonly found in DCNs, causing frequent TCP incast collapses. Second, the syndrome we discover for the first time, results in 2-5 times lower throughput for single-path flows than multipath flows, thus severely violating network fairness.To effectively tackle these problems, we propose AMP: an adaptive multipath congestion control mechanism that quickly detects the onset of these problems and transforms its multipath flow into a single-path flow. Once these problems disappear, AMP safely reverses this transformation and continues its data transmission via multiple paths. Our evaluation results under a diverse set of scenarios in a fat-tree topology with realistic workloads demonstrate that AMP is robust to the TCP incast problem and improves network fairness between multipath and single-path flows significantly with little performance loss.

show abstract

“…We select DPDK (Data Plane Development Kit) as the D-Plane which has been reported to realize high throughput, low latency [6] and low processing jitter [7] in forwarding packets.…”

Section: Separation Of the Data-plane And Control-planementioning

confidence: 99%

Implementation of software-based EPON-OLT and performance evaluation

et al. 2017

View full text Add to dashboard Cite

In order to meet the diversifying demands, we propose the architecture of the software-based virtual EPON-OLT, where all of OLT functions except PHY-Layer are implemented as software running on an x86 server. The EPON-OLT requires not only high throughput, low latency but also accurate controls for the upstream transmission timing of ONUs. In order to satisfy these conditions, we introduced two methods for the virtual OLT: (1) Separation of D-Plane/C-Plane and (2)

show abstract

DX: Latency-Based Congestion Control for Datacenters

Cited by 71 publications

References 28 publications

Timely

Timely

AMP: an adaptive multipath TCP for data center networks

Implementation of software-based EPON-OLT and performance evaluation

Contact Info

Product

Resources

About