Improving resource utilization by timely fine-grained scheduling

Jin, Tatiana; Cai, Zhenkun; Li, Boyang; Zheng, Chengguang; Jiang, Guanxian; Cheng, James

doi:10.1145/3342195.3387551

Cited by 20 publications

(3 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A single scheduler is adopted in Borg [8] to improve the resource utilization with the consideration of machine sharing and performance isolation. Jin et al [20] proposed a task execution framework called Ursa, in which a centralized task scheduler was adopted to assign tasks globally. Although better deployment quality can be obtained, the deployment latency may increase since information about all tasks and hosts needs to be collected, especially when the number of tasks is very large.…”

Section: Scheduler Structuresmentioning

confidence: 99%

Efficient Dynamic Deployment of Simulation Tasks in Collaborative Cloud and Edge Environments

et al. 2022

View full text Add to dashboard Cite

Cloud computing has been studied and used extensively in many scenarios for its nearly unlimited resources and X as a service model. To reduce the latency for accessing the remote cloud data centers, small data centers or cloudlets are deployed near end-users, which is also called edge computing. In this paper, we mainly focus on the efficient scheduling of distributed simulation tasks in collaborative cloud and edge environments. Since simulation tasks are usually tightly coupled with each other by sending many messages and the status of tasks and hosts may also change frequently, it is essentially a dynamic bin-packing problem. Unfortunately, popular methods, such as meta-heuristics, and accurate algorithms are time-consuming and cannot deal with the dynamic changes of tasks and hosts efficiently. In this paper, we present Pool, an incremental flow-based scheduler, to minimize the overall communication cost of all tasks in a reasonable time span with the consideration of migration cost of task. After formulating such a scheduling problem as a min-cost max-flow (MCMF) problem, incremental MCMF algorithms are adopted to accelerate the procedure of calculating an optimal flow and heuristic scheduling algorithm, with the awareness of task migration cost, designed to assign tasks. Simulation experiments on Alibaba cluster trace show that Pool can schedule all of the tasks efficiently and is almost 5.8 times faster than the baseline method when few tasks and hosts change in the small problem scale.

show abstract

Section: Scheduler Structuresmentioning

confidence: 99%

Efficient Dynamic Deployment of Simulation Tasks in Collaborative Cloud and Edge Environments

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Resource over-commitment and virtual machine consolidation in cloud. Overcommitment [10,21] in which the sum of allocated resources to multiple containers is larger than physical capacity, is proposed to increase the resource utilization in recent cloud architecture like Borg [24]. With proper consolidation of multiple containers [23], statistical multiplexing to over-commit resources can avoid almost all resource violations.…”

Section: Related Workmentioning

confidence: 99%

Solving the Batch Stochastic Bin Packing Problem in Cloud: A Chance-constrained Optimization Approach

Yan,

Lu,

Chen

et al. 2022

Preprint

View full text Add to dashboard Cite

This paper investigates a critical resource allocation problem in the first party cloud: scheduling containers to machines. There are tens of services and each service runs a set of homogeneous containers with dynamic resource usage; containers of a service are scheduled daily in a batch fashion. This problem can be naturally formulated as Stochastic Bin Packing Problem (SBPP). However, traditional SBPP research often focuses on cases of empty machines, whose objective, i.e., to minimize the number of used machines, is not well-defined for the more common reality with nonempty machines. This paper aims to close this gap. First, we define a new objective metric, Used Capacity at Confidence (UCaC), which measures the maximum used resources at a probability and is proved to be consistent for both empty and nonempty machines, and reformulate the SBPP under chance constraints. Second, by modeling the container resource usage distribution in a generative approach, we reveal that UCaC can be approximated with Gaussian, which is verified by trace data of real-world applications. Third, we propose an exact solver by solving the equivalent cutting stock variant as well as two heuristics-based solvers -UCaC best fit, bi-level heuristics. We experimentally evaluate these solvers on both synthetic datasets and real application traces, demonstrating our methodology's advantage over traditional SBPP optimal solver minimizing the number of used machines, with a low rate of resource violations. CCS CONCEPTS• Applied computing → Decision analysis.

show abstract

“…The most important task is to meet the clients' expectations through the guarantee of meeting the deadline of the production orders' realization with the simultaneous profit maximization gained through selling the productions orders received for realization (Jardzioch & Skobiej, 2013;Lee & Kim, 2001). The simultaneous goal of scheduling is maximization of utilization the machines, which is more widely described in the manuscripts (Jin et al, 2020;Wang et al, 2020), where the authors discuss the topic of improving resource utilization by timely finegrained scheduling and detailed planning, as well as the problem of multi-objective production scheduling for balanced production of manufactures. The issue of scheduling the production tasks is a significantly complex area analyzed in many research centers.…”

Section: Introductionmentioning

confidence: 99%

Management and Production Engineering Review

Matuszny

2022

View full text Add to dashboard Cite

The article is to present the application of genetic algorithm in production scheduling in a production company. In the research work the assumptions of the methodology were described and the operation of the proposed genetic algorithm was presented in details. Genetic algorithms are useful in complex large scale combinatorial optimisation tasks and in the engineering tasks with numerous limitations in the production engineering. Moreover, they are more reliable than the existing direct search algorithms. The research is focused on the effectivity improvement and on the methodology of scheduling of a manufacturing cell work. The genetic algorithm used in the work appeared to be robust and fast in finding accurate solutions. It was shown by experiment that using this method enables obtaining schedules suitable for a model. It gives a group of solutions that are at least as good as those created by the heuristic rules.

show abstract

Improving resource utilization by timely fine-grained scheduling

Cited by 20 publications

References 28 publications

Efficient Dynamic Deployment of Simulation Tasks in Collaborative Cloud and Edge Environments

Efficient Dynamic Deployment of Simulation Tasks in Collaborative Cloud and Edge Environments

Solving the Batch Stochastic Bin Packing Problem in Cloud: A Chance-constrained Optimization Approach

Management and Production Engineering Review

Contact Info

Product

Resources

About