Abstract:In modern computing systems, jobs' resource requirements often vary over time. Accounting for this temporal variability during job scheduling is essential for meeting performance goals. However, theoretical understanding on how to schedule jobs with time-varying resource requirements is limited. Motivated by this gap, we propose a \emph{new setting} of the stochastic bin-packing problem in service systems that allows for \emph{time-varying} job resource requirements, also referred to as `item sizes' in traditional bin-packing terms. In this setting, a job or `item' must be dispatched to a server or `bin' upon arrival. Its resource requirement may vary over time while in service, following a Markovian assumption. Once the job's service is complete, it departs from the system. Our goal is to minimize the expected number of active servers, or `non-empty bins', in steady state. Under our problem formulation, we develop a job dispatch policy, named Join-Reqesting-Server (JRS). Broadly, JRS lets each server independently evaluate its current job configuration and decide whether to accept additional jobs, balancing the competing objectives of maximizing throughput and minimizing the risk of resource capacity overruns. The JRS dispatcher then utilizes these individual evaluations to decide which server to dispatch each arriving job to. The theoretical performance guarantee of JRS is in the asymptotic regime where the job arrival rate scales large linearly with respect to a scaling factor $r$. We show that JRS achieves an additive optimality gap of $O(\sqrt{r})$ in the objective value, where the optimal objective value is $\Theta(r)$. When specialized to constant job resource requirements, our result improves upon the state-of-the-art $o(r)$ optimality gap. Our technical approach highlights a novel policy conversion framework that reduces the policy design problem into a single-server problem.

Simple Near-Optimal Scheduling for the M/G/1

Optimal Scheduling and Exact Response Time Analysis for Multistage Jobs

Optimal Online Algorithms for Scheduling on Two Identical Machines under a Grade of Service

Performance of the Gittins Policy in the G/G/1 and G/G/k, With and Without Setup Times

How to Schedule Near-Optimally under Real-World Constraints

On-Line Scheduling With Extendable Working Time On A Small Number Of Machines

A Note on On-Line Scheduling with Partial Information

Optimal Scheduling in the Multiserver-job Model under Heavy Traffic

Optimal On-Line Algorithms for Scheduling on Parallel Batch Processing Machines

Strongly Tail-Optimal Scheduling in the Light-Tailed M/G/1

On Resolving Non-Preemptivity in Multitask Scheduling: An Optimal Algorithm in Deterministic and Stochastic Worlds

A C/μ ‐rule for Job Assignment in Heterogeneous Group‐Server Queues

The expected asymptotical ratio for preemptive stochastic online problem

Stochastic scheduling problem with varying weight for each job

Optimal Policy For Single-Machine Scheduling With Deterioration Effects, Learning Effects, Setup Times, And Availability Constraints

Approximate Optimal Solution for a D-Policy in an M/G/X Queuing System

Balanced Splitting: A Framework for Achieving Zero-wait in the Multiserver-job Model

SOAP: One Clean Analysis of All Age-Based Scheduling Policies

A Model for Minimizing Active Processor Time

Near-Optimal Stochastic Bin-Packing in Large Service Systems with Time-Varying Item Sizes