MATE: multipath adaptive traffic engineering

Elwalid, A.; Jin, Cheng; Low, Steven H.; Widjaja, Indra

doi:10.1016/s1389-1286(02)00308-0

Cited by 52 publications

(58 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, because the endpoints of the coflows are fixed (e.g., source and destination specified by location of input data and tasks), these cannot schedule around network bottlenecks. 4) Scheduling on the WAN: There has been much work on optimizing WAN transfers including tuning ECMP weights [32] and adapting allocations across pre-established tunnels [31,39]. Also, both Google [38] and Microsoft [36] recently published details on their production WAN networks.…”

Section: ) Distributed Databasesmentioning

confidence: 99%

Low Latency Geo-distributed Data Analytics

Ananthanarayanan

Bodík

et al. 2015

Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication

194

176

View full text Add to dashboard Cite

Low latency analytics on geographically distributed datasets (across datacenters, edge clusters) is an upcoming and increasingly important challenge. The dominant approach of aggregating all the data to a single datacenter significantly inflates the timeliness of analytics. At the same time, running queries over geo-distributed inputs using the current intra-DC analytics frameworks also leads to high query response times because these frameworks cannot cope with the relatively low and variable capacity of WAN links.We present Iridium, a system for low latency geo-distributed analytics. Iridium achieves low query response times by optimizing placement of both data and tasks of the queries. The joint data and task placement optimization, however, is intractable. Therefore, Iridium uses an online heuristic to redistribute datasets among the sites prior to queries' arrivals, and places the tasks to reduce network bottlenecks during the query's execution. Finally, it also contains a knob to budget WAN usage. Evaluation across eight worldwide EC2 regions using production queries show that Iridium speeds up queries by 3× − 19× and lowers WAN usage by 15% − 64% compared to existing baselines.

show abstract

Section: ) Distributed Databasesmentioning

confidence: 99%

Low Latency Geo-distributed Data Analytics

Ananthanarayanan

Bodík

et al. 2015

Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication

194

176

View full text Add to dashboard Cite

show abstract

“…For achieving these goals, we focus here on the effective mapping of traffic demands onto the network topology and the adaptive reconfiguration of the mapping according to traffic fluctuations. 2 Some studies solve these issues with modified traditional IP routing protocols 3 : however, these solutions have limitations such as difficulty in dealing with QoS provisioning for a variety of traffic.…”

Section: Introduction Traffic Engineering (Te)mentioning

confidence: 99%

Analysis of two-phase path management scheme for MPLS traffic engineering

Tamura

Kawahara

Oie

2004

Performance, Quality of Service, and Control of Next-Generation Communication Networks II

View full text Add to dashboard Cite

Multi-Protocol Label Switching (MPLS) is suitable for implementing Traffic Engineering (TE) to achieve two goals: Quality of Service (QoS) provisioning and the efficient use of network resources. In fact, MPLS allows several detour paths to be (pre-)established for some source-destination pair as well as its primary path of minimum hops. Thus, we focus on a two-phase path management scheme using these two kinds of paths. In the first phase, each primary path is allocated to a flow on a specific source-destination pair if the path is not congested, i.e., if its utilization is less than some predetermined threshold; otherwise, as the second phase, one of the detour paths is allocated randomly if it is available. Therefore, in this paper, we analytically evaluate this path management scheme by extending the M/M/c/c queueing system and we investigate the impact of a threshold on the flow-blocking probability. Through some numerical results, we discuss the adequacy of the path management scheme for MPLS-TE.

show abstract

“…We show convergence to the optimal routing solutions for synchronous versions of the algorithms, with perfect (noiseless) estimates of the queueing delays (these essentially are the convergence results of the corresponding subgradient algorithms). Our optimal routing solution is not an end-to-end solution (our formulation is not a path-based formulation) like many of the above-cited works [29], [18], [30]. Consequently, our algorithms would avoid the scalability issues related to such an approach.…”

mentioning

confidence: 99%

“…The relations (17), (18), and (19), can be used as a basis for a distributed algorithm that converges to a solution of the dual optimization problem. We now describe a general, online version of such an algorithm.…”

mentioning

confidence: 99%

“…Although we have described a general, asynchronous, distributed algorithm for the dual problem, we shall discuss convergence only for the special case of the synchronous version as given by equations (17), (18), and (19), with 'perfect' ('noiseless') measurements of the delays. We can view the gradient algorithm (15) as a special case of a subgradient algorithm, and employ the results of convergence analysis (see, for example, Shor [28]) for the latter algorithm.…”

mentioning

confidence: 99%

See 1 more Smart Citation

An Optimal Distributed Routing Algorithm using Dual Decomposition Techniques

Purkayastha¹,

Baras²

2008

Communications in Information and Systems

View full text Add to dashboard Cite

Abstract. We consider the routing problem in wireline, packet-switched communication networks. We cast our optimal routing problem in a multicommodity network flow optimization framework. Our cost function is related to the congestion in the network, and is a function of the flows on the links of the network. The optimization is over the set of flows in the links corresponding to the various destinations of the incoming traffic. We separately address the single commodity and the multicommodity versions of the routing problem. We consider the dual problems, and using dual decomposition techniques, we provide primal-dual algorithms that converge to the optimal solutions of the problems. Our algorithms, which are subgradient algorithms to solve the corresponding dual problems, can be implemented in a distributed manner by the nodes of the network. For online, adaptive implementations of our algorithms, the nodes in the network need to utilize 'locally available information' like estimates of queue lengths on outgoing links. We show convergence to the optimal routing solutions for synchronous versions of the algorithms, with perfect (noiseless) estimates of the queueing delays. Every node of the network controls the flows on the outgoing links using the distributed algorithms. For both the single commodity and multicommodity cases, we show that our algorithm converges to a loop-free optimal solution. Our optimal solutions also have the attractive property of being multipath routing solutions.

show abstract

MATE: multipath adaptive traffic engineering

Cited by 52 publications

References 13 publications

Low Latency Geo-distributed Data Analytics

Low Latency Geo-distributed Data Analytics

Analysis of two-phase path management scheme for MPLS traffic engineering

An Optimal Distributed Routing Algorithm using Dual Decomposition Techniques

Contact Info

Product

Resources

About