On the Throughput Optimization in Large-Scale Batch-Processing Systems

@article{Kar2020OnTT,
  title={On the Throughput Optimization in Large-Scale Batch-Processing Systems},
  author={Sounak Kar and Robin Rehrmann and Arpan Mukhopadhyay and Bastian Alt and Florin Ciucu and Heinz Koeppl and Carsten Binnig and Amr Rizk},
  journal={ArXiv},
  year={2020},
  volume={abs/2009.09433}
}
We analyze a data-processing system with $n$ clients producing jobs which are processed in \textit{batches} by $m$ parallel servers; the system throughput critically depends on the batch size and a corresponding sub-additive speedup function. In practice, throughput optimization relies on numerical searches for the optimal batch size, a process that can take up to multiple days in existing commercial systems. In this paper, we model the system in terms of a closed queueing network; a standard… Expand
On the Throughput Optimization in Large-Scale Batch-Processing Systems
TLDR
A mean-field model is proposed that has a unique, globally attractive stationary point, derivable in closed form that characterizes the asymptotic throughput as a function of the batch size that can be calculated in O(1) time. Expand
Discrete-Time Analysis of Wireless Blockchain Networks
TLDR
This paper proposes a discrete-time Markov model to capture the expected delay incurred by the BC when the supporting communication platform is wireless, specifically on IEEE 802.11ax, for the use case of users’ radio resource provisioning. Expand

References

SHOWING 1-10 OF 50 REFERENCES
Randomized Assignment of Jobs to Servers in Heterogeneous Clusters of Shared Servers for Low Delay
TLDR
This work considers the job assignment problem in a multi-server system consisting of parallel processor sharing servers, categorized into different types according to their processing capacity or speed and establishes the existence and uniqueness of stationary solution of the mean field. Expand
Analysis of finite-buffer state-dependent bulk queues
TLDR
This paper uses semi-regenerative analysis to develop a numerically stable method for calculating the limiting probability distribution of the queue length process, and presents various performance measures for evaluating admission control and batch service policies. Expand
The Power of Two Choices in Randomized Load Balancing
TLDR
This work uses a limiting, deterministic model representing the behavior as n/spl rarr//spl infin/ to approximate the behavior of finite systems and provides simulations that demonstrate that the method accurately predicts system behavior, even for relatively small systems. Expand
Analysis of Randomized Join-the-Shortest-Queue (JSQ) Schemes in Large Heterogeneous Processor-Sharing Systems
TLDR
It is shown that it is possible to combine the static state-independent scheme with a randomized JSQ scheme that allows us to recover the maximal stability region combined with the benefits of JSQ, and such a scheme is preferable in terms of average delay. Expand
Power of d Choices for Large-Scale Bin Packing: A Loss Model
TLDR
This paper considers power-of-d-choices routing, where a job is routed to the server with the largest amount of available resource among d ≥ 2 randomly chosen servers, and considers a fluid model that corresponds to the limit as N goes to infinity and provides an explicit upper bound for the equilibrium blocking probability. Expand
Closed Queuing Systems with Exponential Servers
TLDR
It is found that the distribution of customers in the closed queuing system is regulated by the stage or stages with the slowest effective service rate, which means that closed systems are shown to be stochastically equivalent to open systems in which the number of customers cannot exceed N. Expand
OPTIMAL CONTROL OF BATCH SERVICE QUEUES
A batch service queue is considered where each batch size and its time of service is subject to control. Costs are incurred for serving the customers and for holding them in the system. Viewing theExpand
Open, Closed, and Mixed Networks of Queues with Different Classes of Customers
TLDR
Many of the network results of Jackson on arrival and service rate dependencies, of Posner and Bernholtz on different classes of customers, and of Chandy on different types of service centers are combined and extended in this paper. Expand
Optimal Batch Provisioning to Customers Subject to a Delay-Limit
This work deals with batch provisioning and order aggregation. Two examples are: (i) a manufacturer that has to deliver items to customers in a remote destination, and (ii) a company that providesExpand
Optimal Dispatching of a Finite Capacity Shuttle.
We consider the problem of determining the optimal operating policy of a two terminal shuttle with fixed capacity Q \le \infty . The passengers arrive at each terminal according to Poisson processesExpand
...
1
2
3
4
5
...