An Online Virtual Machine Placement Algorithm in an Over-Committed Cloud

Ji, Siqi; Li, Ming Da; Ji, Niannian; Li, Baochun

doi:10.1109/ic2e.2018.00033

Cited by 6 publications

(6 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, Tetris ignored the temporal resource usage changes during runtime and was designed under assumption of priori knowledge on both the resource requirements of tasks and resource availability at machines. Authors in [19] considered balanced usage of multiple dimensional resources to design a threshold-based VM placement strategy Min-DIFF, which is intended to balance the usage of resources and reduce the risk of PM overloading in an online manner but the threshold of each dimensional resource didn't vary with time-varying recourse requirements and resource usage.…”

Section: Related Workmentioning

confidence: 99%

“…Google Trace. We use real data center workload traces from Google cluster-usage traces [38], [39] to emulate the multi-dimension resource demands of traffic, similar to the datasets in [22] and [19]. Google trace records the resource usage data of different workload in a Google data center over a month-long period, which contains the CPU, memory and disk requirements (normalized by the resource of one server) from workload, as well as arrival time (absolute time value) and durations.…”

Section: Multi-dimensional Resource Utilizationmentioning

confidence: 99%

“…We evaluate the proposed NESRL-MRM and compare its performance against four baseline schemes, including a DRL-based resource allocation scheme according to A2CScheduler in [27], a threshold-based algorithm Min-DIFF in [19], a tight packing heuristic algorithm (termed as TPA), and a traditional round-robin resource allocation policy (denoted RRA).…”

Section: B Baseline Schemesmentioning

confidence: 99%

“…As one of the most related virtual machines placement problem in cloud DCs, a threshold-based algorithm, termed as Min-DIFF [19], is used as a baseline scheme.…”

Section: B Baseline Schemesmentioning

confidence: 99%

See 3 more Smart Citations

Multi-Dimensional Resource Allocation in Distributed Data Centers Using Deep Reinforcement Learning

Wei

Wang

et al. 2023

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

With the development of edge-cloud computing technologies, distributed data centers (DCs) have been extensively deployed across the global Internet. Since different users/applications have heterogeneous requirements on specific types of ICT resources in distributed DCs, how to optimize such heterogeneous resources under dynamic and even uncertain environments becomes a challenging issue. Traditional approaches are not able to provide effective solutions for multi-dimensional resource allocation that involves the balanced utilization across different resource types in distributed DC environments. This paper presents a reinforcement learning based approach for multidimensional resource allocation (termed as NESRL-MRM) that is able to achieve balanced utilization and availability of resources in dynamic environments. To train NESRL-MRM's agent with sufficiently quick wall-clock time but without the loss of exploration diversity in the search space, a natural evolution strategy (NES) is employed to approximate the gradient of the reward function. To realistically evaluate the performance of NESRL-MRM, our simulation evaluations are based on real-world workload traces from Amazon EC2 and Google datacenters. Our results show that NESRL-MRM is able to achieve significant improvement over the existing approaches in balancing the utilization of multi-dimensional DC resources, which leads to substantially reduced blocking probability of future incoming workload demands.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Multi-dimensional Resource Utilizationmentioning

confidence: 99%

Section: B Baseline Schemesmentioning

confidence: 99%

“…As one of the most related virtual machines placement problem in cloud DCs, a threshold-based algorithm, termed as Min-DIFF [19], is used as a baseline scheme.…”

Section: B Baseline Schemesmentioning

confidence: 99%

See 2 more Smart Citations

Multi-Dimensional Resource Allocation in Distributed Data Centers Using Deep Reinforcement Learning

Wei

Wang

et al. 2023

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

show abstract

“…For example, in MapReduce, an established framework for processing large-scale dataintensive applications [8], big data is partitioned and stored over several data nodes in a cloud system [9], and the model undertakes efficient parallel computing through a large number of data nodes and computation nodes for dataintensive cloud application [10]. Hence, improper virtual machine (VM) placement can cause unbalanced resource utilization [11], transmission latency [5], [9][10], increasing bandwidth usage [9], and energy consumption [12][13]. All of these challenges increase service costs and violate service level agreements (SLA) from a business perspective.…”

Section: Introductionmentioning

confidence: 99%

Virtual Machine Placement Optimization for Big Data Applications in Cloud Computing

Seyyedsalehi

Khansari

2022

IEEE Access

View full text Add to dashboard Cite

Big data and cloud computing are two advanced technologies that have overcome many computing and analytical challenges in recent years. With the rise in the applications of these technologies, the necessity of efficiency and optimization in the utilization of related resources has made sense. The procedure of locating virtual machines (VM) in physical machines (PM) affects the performance, speed, and costs of cloud computing services. VM placement in cloud computing is an NP-hard problem. Indeed, the problem is more complicated in big data tasks due to the need for transferring high volumes of traffic between VMs. This paper proposes a new approach for VM placement in a multi data center (DC) cloud environment. The aware genetic algorithm first fit (AGAFF) is a context-aware algorithm that distinguishes big data tasks with an input tag and uses a structure to minimize the traffic between MapReduce nodes. This multi-objective algorithm is based on the genetic algorithm, which is incorporated with the first fit methodology. The algorithm minimizes energy usage by minimizing the number of used servers, intra-DC traffic of big data tasks, and VMs' live migration while maximizing relevant usage of CPU and RAM in every server. Furthermore, it improves job execution time, especially in big data processing, and reduces service level agreement (SLA) violations. A comparison between the results of AGAFF and four other algorithms shows by about 61% energy consumption reduction on average on different scales and approves a decrease in the number of needed PMs, intra-DC traffic of big data processing, and the number of live migrations.

show abstract

Scheduling Virtual Machine Migration During Datacenter Upgrades with Reinforcement Learning

Chen

et al. 2020

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

An Online Virtual Machine Placement Algorithm in an Over-Committed Cloud

Cited by 6 publications

References 23 publications

Multi-Dimensional Resource Allocation in Distributed Data Centers Using Deep Reinforcement Learning

Multi-Dimensional Resource Allocation in Distributed Data Centers Using Deep Reinforcement Learning

Virtual Machine Placement Optimization for Big Data Applications in Cloud Computing

Scheduling Virtual Machine Migration During Datacenter Upgrades with Reinforcement Learning

Contact Info

Product

Resources

About