Light traffic performance under the power of two load balancing strategy

Izagirre, Ane; Makowski, Armand M.

doi:10.1145/2667522.2667527

Cited by 13 publications

(16 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For that, one needs to find the extrema of f 1 (λ)/f 2 (λ) in (0,μ). We conjecture that B is close to B 0 defined in (18). In fact, numerically, it seems that 1 − V 2 V 1 > 2B 0 already ensures that for all λ ∈ (0,μ), N 1 (λ) N 2 (λ).…”

Section: A2 Additional Remarks On the Boundsmentioning

confidence: 53%

“…Packets arrive in the system as a Poisson process with rate parameter λ, and are routed to the queues based on a Bernoulli trial: With probability p, a packet goes to Queue 1, with probability 1 − p, it goes to Queue 2. As opposed to most works that study queues in parallel [14,17,18], we assume that the routing decision is made without knowledge of the current size of the queues. Several reasons justify this choice.…”

Section: Model For Simultaneous Pathsmentioning

confidence: 99%

“…We use the following values, with the service rates (µ) in packets/s, and the transition rates These values are chosen randomly among some for which the smaller delays are achieved by the queue with larger variance for certain arrival rates, as shown by Corollary 3.4. More precisely, we choose values such that (18) in Appendix is verified. We havê µ 1 =μ 2 = 38.4 Mb/s, V 1 = 1.33 × 10 6 , and V 2 = 1.67 × 10 6 : Queue 1 is the queue with smallest variance.…”

Section: Larger Variance Can Imply Smaller Delaysmentioning

confidence: 99%

See 2 more Smart Citations

On the Delays in Time-Varying Networks

Henri

Shneer

Thiran

2018

Proceedings of the Eighteenth ACM International Symposium on Mobile Ad Hoc Networking and Computing

View full text Add to dashboard Cite

In all networks, link or route capacities fluctuate for multiple reasons, e.g., fading and multi-path effects on wireless channels, interference and contending users on a shared medium, varying loads in WAN routers, impedance changes on power-line channels. These fluctuations severely impact packet delays. In this paper, we study delays in time-varying networks. Intuitively, we expect that for a given average service rate, an increased service rate variability yields larger delays. We find that this is not always the case. Using a queuing model that includes time-varying service rates, we show that for certain arrival rates, a queue with larger service rate variance offers smaller average delays than a queue with the same average service rate and lower service rate variance. We also verify these findings on a wireless testbed. We then study the conditions under which using simultaneously two independent paths helps in terms of delays, for example, in hybrid networks where the two paths use different physical layer technologies. We show that using two paths is not always better, in particular for low arrival rates. We also show that the optimal traffic splitting between the two paths depends on the arrival rate.

show abstract

Section: A2 Additional Remarks On the Boundsmentioning

confidence: 53%

Section: Model For Simultaneous Pathsmentioning

confidence: 99%

Section: Larger Variance Can Imply Smaller Delaysmentioning

confidence: 99%

See 1 more Smart Citation

On the Delays in Time-Varying Networks

Henri

Shneer

Thiran

2018

Proceedings of the Eighteenth ACM International Symposium on Mobile Ad Hoc Networking and Computing

View full text Add to dashboard Cite

show abstract

“…In particular it ensures the requisite differentiability of λ → R(λ) where finite. We compute the first two derivatives in light traffic; these results were announced in the conference paper (Izagirre and Makowski, 2014) without proofs.…”

Section: Evaluating the First Two Derivativesmentioning

confidence: 99%

Light traffic behavior under the power-of-two load balancing strategy: The case of heterogeneous servers

Izagirre

Makowski

2017

Performance Evaluation

Self Cite

View full text Add to dashboard Cite

We consider a multi-server queueing system under the power-of-two policy with Poisson job arrivals, heterogeneous servers and a general job requirement distribution; each server operates under the firstcome first-serve policy and there are no buffer constraints. We analyze the performance of this system in light traffic by evaluating the first two light traffic derivatives of the average job response time. These expressions point to several interesting structural features associated with server heterogeneity in light traffic: For unequal capacities, the average job response time is seen to decrease for small values of the arrival rate, and the more diverse the server speeds, the greater the gain in performance. These theoretical findings are assessed through limited simulations.

show abstract

“…The corresponding case of the randomized longest-queue-first policy was addressed asymptotically in Alanyali and Dashouk [2] and Dieker and Suk [5]. The case of heterogenous servers, Poisson arrivals and general service time distributions was addressed under an Erlang loss server model by Mukhopadhyay et al [14] and also under a light load regime with FCFS servers by Izagirre and Makowski [7].…”

Section: Introductionmentioning

confidence: 99%