MMPTCP: A multipath transport protocol for data centers

Kheirkhah, Morteza; Wakeman, Ian; Parisis, George

doi:10.1109/infocom.2016.7524530

Cited by 72 publications

(55 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We use permutation traffic matrix that has been used in many previous works [3,10,37,12,25,16]. Specifically, a host establishes at most two connections: one for receiving traffic and the other for sending traffic.…”

Section: Large-scale Simulationmentioning

confidence: 99%

“…A large-scale data center comprises hundreds of thousands of servers, and hosts a diverse set of online services that require high bandwidth, low latency or both from the network. To meet such requirements, lots of recent advances [5,32,37,28,10,11,40,34,30,44,29,46,45] have focused on improving TCP congestion control (CC) algorithms, by leveraging path diversity [37,28,10,11], exploiting explicit congestion signals from network switches [5,10,40,45,34], measuring delays [29,32], etc.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

AMP: an adaptive multipath TCP for data center networks

Kheirkhah

Lee

2019

2019 IFIP Networking Conference (IFIP Networking)

Self Cite

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

Section: Large-scale Simulationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

AMP: an adaptive multipath TCP for data center networks

Kheirkhah

Lee

2019

2019 IFIP Networking Conference (IFIP Networking)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Several examples of multipath transports include MPTCP [140], XMP [183] and MMPTCP [184]. MPTCP leverages ECMP to route sub-flows over various paths and increase total throughput while balancing load across paths by moving load across sub-flows.…”

Section: E Multipathingmentioning

confidence: 99%

Datacenter Traffic Control: Understanding Techniques and Trade-offs

Noormohammadpour¹,

Raghavendra²

2017

Preprint

View full text Add to dashboard Cite

Datacenters provide cost-effective and flexible access to scalable compute and storage resources necessary for today's cloud computing needs. A typical datacenter is made up of thousands of servers connected with a large network and usually managed by one operator. To provide quality access to the variety of applications and services hosted on datacenters and maximize performance, it deems necessary to use datacenter networks effectively and efficiently. Datacenter traffic is often a mix of several classes with different priorities and requirements. This includes user-generated interactive traffic, traffic with deadlines, and long-running traffic. To this end, custom transport protocols and traffic management techniques have been developed to improve datacenter network performance.In this tutorial paper, we review the general architecture of datacenter networks, various topologies proposed for them, their traffic properties, general traffic control challenges in datacenters and general traffic control objectives. The purpose of this paper is to bring out the important characteristics of traffic control in datacenters and not to survey all existing solutions (as it is virtually impossible due to massive body of existing research). We hope to provide readers with a wide range of options and factors while considering a variety of traffic control mechanisms. We discuss various characteristics of datacenter traffic control including management schemes, transmission control, traffic shaping, prioritization, load balancing, multipathing, and traffic scheduling. Next, we point to several open challenges as well as new and interesting networking paradigms. At the end of this paper, we briefly review inter-datacenter networks that connect geographically dispersed datacenters which have been receiving increasing attention recently and pose interesting and novel research problems.

show abstract

“…Rural areas are, however, prone to having infrastructure that does not support fast network bandwidth and reliable communication paths, which, resultantly, hinders the implementation of efficacious telehealth services. Multipath transmission schemes may be able to circumvent these issues as, unlike single‐path transmission schemes, they capitalise on the bandwidth resource of several paths to increase throughput, thus segueing around congested paths …”

Section: Introductionmentioning

confidence: 99%

“…Multipath transmission schemes may be able to circumvent these issues as, unlike single-path transmission schemes, they capitalise on the bandwidth resource of several paths to increase throughput, thus segueing around congested paths. [5][6][7] Recently, Wu et al 8 outlined two major challenges that affect real-time video streaming from a single sender (server) to a multihomed mobile device: bandwidth efficiency and end-to-end delay. They sought to address those issues by developing a framework that uses Video and Raptor code-aware concurrent multipath transfer (CMT-VR; see Figure 1).…”

Section: Introductionmentioning

confidence: 99%

A novel solution for real‐time wireless networks: Telehealth psychological services in rural area

Luitel

Alsadoon

Prasad

et al. 2019

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Telehealth psychological services is a time‐efficient and cost‐effective therapy solution for persons in rural areas. The high congestion resulting from these infrastructure issues, which includes low‐bandwidth wireless connections and channels prone to data loss, tends to increase video delays, thereby making real‐time video transmission during telehealth psychological services challenging. The purpose of this study is, hence, to enable the successful implementation of telehealth psychological services in rural areas by minimising end‐to‐end video transmission delay and improving bandwidth efficiency. Our proposed system uses a distortion‐aware TCP‐friendly adaptive congestion control algorithm that adapts the congestion window size according to path characteristics and quality. Indeed, evaluation results show that our algorithm—as compared to a reference—significantly improves peak signal‐to‐noise ratio and end‐to‐end delay; thus, suggesting that the adoption of an adaptive congestion window size in our system can improve network bandwidth efficiency and reduce information delay resulting from packet queuing and unnecessary packet loss. The outcome is successful high‐definition video transmission during telehealth psychological services.

show abstract

MMPTCP: A multipath transport protocol for data centers

Cited by 72 publications

References 17 publications

AMP: an adaptive multipath TCP for data center networks

AMP: an adaptive multipath TCP for data center networks

Datacenter Traffic Control: Understanding Techniques and Trade-offs

A novel solution for real‐time wireless networks: Telehealth psychological services in rural area

Contact Info

Product

Resources

About