Mechanisms for Optimizing Link Aggregation Group (LAG) and Equal-Cost Multipath (ECMP) Component Link Utilization in Networks

Khasnabish, Bhumip; Krishnan, Ram; Ghanwani, Anoop; Yong, Lucy

doi:10.17487/rfc7424

Cited by 10 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For comparing the ability of distinct sampling techniques in assisting network flow analysis correctly, several flow parameters are considered, namely: (i) the amount of flows identified; (ii) the percentage of heavy-hitter (HH) flows identified, where the notion of heavy hitter refers to 20% of the largest flows in terms of size (number of packets) [13]; (iii) the utilization share at transport level; (iv) the utilization share at application level; and, (v) the accuracy of load estimations for the identified flows.…”

Section: B Comparative Parametersmentioning

confidence: 99%

Analysing traffic flows through sampling: A comparative study

Silva

Carvalho

Lima

2015

2015 IEEE Symposium on Computers and Communication (ISCC)

View full text Add to dashboard Cite

Understanding network workload through the characterization of network flows, being essential for assisting network management tasks, can benefit largely from traffic sampling as long as an accurate snapshot of network behavior is captured. This paper is devoted to evaluate the real applicability of using sampling to support flow analysis. Considering both classical and emerging sampling techniques, a comparative performance study is carried out to assess the accuracy of estimating flow parameters through sampling. After identifying the main building blocks of sampled-based measurements, a sampling framework has been implemented to provide a versatile and fair platform for carrying out the testing and comparison process. Through an encompassing coverage of representative sampling techniques, the present study aims to provide useful insights regarding the use of sampling in traffic flow analysis.

show abstract

Section: B Comparative Parametersmentioning

confidence: 99%

Analysing traffic flows through sampling: A comparative study

Silva

Carvalho

Lima

2015

2015 IEEE Symposium on Computers and Communication (ISCC)

View full text Add to dashboard Cite

show abstract

“…On the other hand, all the coupled congestion control strategies are designed for connections between the same endpoints having homogeneous round-trip times, and they tend not to fully leverage the statefulness of the TCP congestion control algorithm. Their mechanisms rely on the assumption that connections between the same endpoints share a common bottleneck -this may be true for routers in theory, but in reality two connections may take different routes if they have different TCP port numbers [8].…”

Section: Introductionmentioning

confidence: 99%

ctrlTCP: Reducing latency through coupled, heterogeneous multi-flow TCP congestion control

Islam

Welzl

Hiorth

et al. 2018

IEEE INFOCOM 2018 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS)

View full text Add to dashboard Cite

We present ctrlTCP, a method to combine the congestion controls of multiple TCP connections. In contrast to the previous methods such as the Congestion Manager, ctrlTCP can couple all TCP flows that leave one sender, traverse a common bottleneck (e.g., a home user's thin uplink) and arrive at different destinations. Using ns-2 simulations and an implementation in the FreeBSD kernel, we show that our mechanism reduces queuing delay, packet loss, and short flow completion times while enabling precise allocation of the share of the available bandwidth between the connections according to the needs of the applications.

show abstract

“…All the aforementioned approaches have an additional, entirely different problem: they assume that multiple TCP connections sending to the same destination would take the same path. This is not always true -load-balancing mechanisms such as Equal-Cost Multi-Path (ECMP) and LAG [6] may force them to take different paths. Therefore, in this particular scenario, combining the congestion controllers would incur wrong behavior.…”

Section: Introductionmentioning

confidence: 99%

Lightweight and flexible single-path congestion control coupling

Islam

Welzl

Gjessing

2018

NOMS 2018 - 2018 IEEE/IFIP Network Operations and Management Symposium

View full text Add to dashboard Cite

Communication between two Internet hosts using parallel connections may result in unwanted interference between the connections. In this dissertation, we propose a senderside solution to address this problem by letting the congestion controllers of the different connections collaborate, correctly taking congestion control logic into account. Real-life experiments and simulations show that our solution works for a wide variety of congestion control mechanisms, provides great flexibility when allocating application traffic to the connections, and results in lower queuing delay and less packet loss.

show abstract

Mechanisms for Optimizing Link Aggregation Group (LAG) and Equal-Cost Multipath (ECMP) Component Link Utilization in Networks

Cited by 10 publications

References 21 publications

Analysing traffic flows through sampling: A comparative study

Analysing traffic flows through sampling: A comparative study

ctrlTCP: Reducing latency through coupled, heterogeneous multi-flow TCP congestion control

Lightweight and flexible single-path congestion control coupling

Contact Info

Product

Resources

About