Scheduling packets over multiple interfaces while respecting user preferences

Yap, Kok-Kiong; Huang, Te-Yuan; Yiakoumis, Yiannis; Chinchali, Sandeep; McKeown, Nick

doi:10.1145/2535372.2535387

Cited by 35 publications

(32 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Scheduling in the presence of placement constraints. There are some recent works investigating max-min fair allocation/scheduling for one type of resource while respecting placement constraints [19], [20], [21], [22], [23]. These single-resource schedulers could be useful in a multi-resource setting when one of the resources is dominantly requested by all users.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

An Efficient and Fair Multi-Resource Allocation Mechanism for Heterogeneous Servers

Khamse-Ashari

Lambadaris

Kesidis

et al. 2018

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

Efficient and fair allocation of multiple types of resources is a crucial objective in a cloud/distributed computing cluster. Users may have diverse resource needs. Furthermore, diversity in server properties/capabilities may mean that only a subset of servers may be usable by a given user. In platforms with such heterogeneity, we identify important limitations in existing multi-resource fair allocation mechanisms, notably Dominant Resource Fairness (DRF) and its followup work. To overcome such limitations, we propose a new server-based approach; each server allocates resources by maximizing a per-server utility function. We propose a specific class of utility functions which, when appropriately parameterized, adjusts the trade-off between efficiency and fairness, and captures a variety of fairness measures (such as our recently proposed Per-Server Dominant Share Fairness). We establish conditions for the proposed mechanism to satisfy certain properties that are generally deemed desirable, e.g., envy-freeness, sharing incentive, bottleneck fairness, and Pareto optimality. To implement our resource allocation mechanism, we develop an iterative algorithm which is shown to be globally convergent. Finally, we show how the proposed mechanism could be implemented in a distributed fashion. We carry out extensive trace-driven simulations to show the enhanced performance of our proposed mechanism over the existing ones.1. A resource-pool may consist of a group of homogeneous servers. 2. The assumption of linearly proportionate resource needs is admittedly an idealization. We are following convention set by DRF and used by follow up works.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Proof of Theorem 5: The proof follows exactly the same line of arguments as Theorem 4. Specifically, let λ i , and ν n,i , denote the Lagrange multipliers corresponding to the constraints in (19), and (20). The Lagrangian function for Problem 2 at server i is given by:…”

Section: Appendixmentioning

confidence: 99%

An Efficient and Fair Multi-Resource Allocation Mechanism for Heterogeneous Servers

Khamse-Ashari

Lambadaris

Kesidis

et al. 2018

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

show abstract

“…There are some recent works investigating max-min fair allocation/scheduling for one type of resource while respecting placement constraints [2], [8], [9], [10], [11], [12]. These schedulers could be useful in a multi-resource setting only when one of the resources serves as the bottleneck for all users, otherwise they might result in poor resource utilization [3], [2].…”

Section: Introductionmentioning

confidence: 99%

Per-Server Dominant-Share Fairness (PS-DSF): A multi-resource fair allocation mechanism for heterogeneous servers

Khamse-Ashari

Lambadaris

Kesidis

et al. 2017

2017 IEEE International Conference on Communications (ICC)

View full text Add to dashboard Cite

Users of cloud computing platforms pose different types of demands for multiple resources on servers (physical or virtual machines). Besides differences in their resource capacities, servers may be additionally heterogeneous in their ability to service users -certain users' tasks may only be serviced by a subset of the servers. We identify important shortcomings in existing multi-resource fair allocation mechanisms -Dominant Resource Fairness (DRF) and its follow up work -when used in such environments. We develop a new fair allocation mechanism called Per-Server Dominant-Share Fairness (PS-DSF) which we show offers all desirable sharing properties that DRF is able to offer in the case of a single "resource pool" (i.e., if the resources of all servers were pooled together into one hypothetical server). We evaluate the performance of PS-DSF through simulations. Our evaluation shows the enhanced efficiency of PS-DSF compared to the existing allocation mechanisms. We argue how our proposed allocation mechanism is applicable in cloud computing networks and especially large scale data-centers.

show abstract

“…This means that smart mobile devices can operate in heterogenous network environments, use different protocols and can now operate over much wider bandwidths. A natural way to tackle the problem of traffic congestion is to take advantage of multiple network interfaces available on these devices by using these interfaces intelligently to distribute data content among the users [3][4][5][6].…”

Section: Introductionmentioning

confidence: 99%

Cooperative content delivery exploiting multiple wireless interfaces: methods, new technological developments, open research issues and a case study

et al. 2015

View full text Add to dashboard Cite

In this tutorial paper, we discuss and compare cooperative content delivery (CCD) techniques that exploit multiple wireless interfaces available on mobile devices to efficiently satisfy the already massive and rapidly growing user demand for content. The discussed CCD techniques include simultaneous use of wireless interfaces, opportunistic use of wireless interfaces, and aggregate use of wireless interfaces. We provide a taxonomy of different ways in which multiple wireless interfaces are exploited for CCD, and also discuss the real measurement studies that evaluate the content delivery performance of different wireless interfaces in terms of energy consumption and throughput. We describe several challenges related to the design of CCD methods using multiple interfaces, and also explain how new technological developments can help in accelerating the performance of such CCD methods. The new technological developments discussed in this paper include wireless interface aggregation, network caching, and the use of crowdsourcing. We provide a case study for selection of devices in a group for CCD using multiple interfaces. We consider this case study based on the observation that in general different CCD users can have different link qualities in terms of transmit/receive performance, and selection of users with good link qualities for CCD can accelerate the content delivery performance of wireless networks. Finally, we discuss some open research issues relating to CCD using multiple interfaces.

show abstract

Scheduling packets over multiple interfaces while respecting user preferences

Cited by 35 publications

References 18 publications

An Efficient and Fair Multi-Resource Allocation Mechanism for Heterogeneous Servers

An Efficient and Fair Multi-Resource Allocation Mechanism for Heterogeneous Servers

Per-Server Dominant-Share Fairness (PS-DSF): A multi-resource fair allocation mechanism for heterogeneous servers

Cooperative content delivery exploiting multiple wireless interfaces: methods, new technological developments, open research issues and a case study

Contact Info

Product

Resources

About