Abstract:In general, obtaining the exact steady-state distribution of queue lengths is not feasible. Therefore, we establish bounds for the tail probabilities of queue lengths. Specifically, we examine queueing systems under Heavy-Traffic (HT) conditions and provide exponentially decaying bounds for the probability $\mathbb P(\epsilon q > x)$, where $\epsilon$ is the HT parameter denoting how far the load is from the maximum allowed load. Our bounds are not limited to asymptotic cases and are applicable even for finite values of $\epsilon$, and they get sharper as $\epsilon \to 0$. Consequently, we derive non-asymptotic convergence rates for the tail probabilities. Unlike other approaches such as moment bounds based on drift arguments and bounds on Wasserstein distance using Stein's method, our method yields sharper tail bounds. Furthermore, our results offer bounds on the exponential rate of decay of the tail, given by $-\frac{1}{x} \log \mathbb P(\epsilon q > x)$ for any finite value of $x$. These can be interpreted as non-asymptotic versions of Large Deviation (LD) results. We demonstrate our approach by presenting tail bounds for: (i) a continuous time Join-the-shortest queue (JSQ) load balancing system, (ii) a discrete time single-server queue and (iii) an $M/M/n$ queue. We not only bridge the gap between classical-HT and LD regimes but also explore the large system HT regimes for JSQ and $M/M/n$ systems. In these regimes, both the system size and the system load increase simultaneously. Our results also close a gap in the existing literature on the limiting distribution of JSQ in the super-NDS (a.k.a. super slowdown) regime. This contribution is of an independent interest. Here, a key ingredient is a more refined characterization of state space collapse for JSQ system, achieved by using an exponential Lyapunov function designed to approximate the $\ell_{\infty}$ norm.

Heavy Tails in Queueing Systems: Impact of Parallelism on Tail Performance

HEAVY TAILS IN QUEUING SYSTEMS: IMPACT OF PARAL- LELISM ON TAIL PERFORMANCE

Can Multipath Mitigate Power Law Delays? - Effects of Parallelism on Tail Performance

Efficient parallel algorithms for tandem queueing system simulation

Exponential Tail Bounds on Queues: A Confluence of Non-Asymptotic Heavy Traffic and Large Deviations

Fairness in overloaded parallel queues

Heavy Loads and Heavy Tails

Toward Optimal Operator Parallelism for Stream Processing Topology with Limited Buffers

Overlap Times in Tandem Queues: Identically Distributed Station Case

Parallel Machines Scheduling with Deteriorating and Learning Effects

Efficient algorithms for tandem queueing system simulation

TAIL PROBABILITIES IN QUEUEING PROCESSES

Heavy Traffic Joint Queue Length Distribution without Resource Pooling

Statistical Age-of-Information Bounds for Parallel Systems: When Do Independent Channels Make a Difference?

Performance Analysis of Parallel Processing Systems with Horizontal Decomposition

Age-Delay Tradeoffs in Queueing Systems

Double-ended queues with non-Poisson inputs and their effective algorithms

Generalized Parallel-Server Fork-Join Queues with Dynamic Task Scheduling.

Taming Tail Latency for Erasure-Coded, Distributee Storage Systems

Modulated Branching Processes, Origins of Power Laws and Queueing Duality

A Theoretical Study of Process Dependence for Critical Statistics in Standard Serial Models and Standard Parallel Models