Autonomic resource provisioning in rocks clusters using Eucalyptus cloud computing

Kijsipongse, Ekasit; Vannarat, Sornthep

doi:10.1145/1936254.1936265

Cited by 9 publications

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The rocks cluster software includes a Xen package for managing the deployment of virtual clusters. In , the authors propose an autonomous approach to achieve transparent integration of virtual resources to extend the capacity of clusters based on the rocks Xen package and the torque cluster manager. Although it was possible to leverage a similar configuration–deployment–contextualization cycle in the context of our simplified characterization, we designed VALPA on libvirt instead in order to explore the VMM configuration parameters, which we believe can impact HPC workloads significantly.…”

Section: Related Studiesmentioning

confidence: 99%

An analysis of definition and placement of virtual machines for high performance applications on Clouds

Evoy

Mury

Schulze

2014

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYThe efficient deployment of high performance computing applications on Clouds offers many challenges, in particular, for communication‐intensive applications. One strategy to mitigate performance overheads caused by high‐communication latency is to schedule requested virtual machines (VMs) effectively onto physical resources by optimizing VM placement. In current approaches based on Infrastructure as a Service resource provisioning, the task of selecting VM profiles is either left to the user or resolved with policies independent from usage patterns, with potential for performance degradation caused by virtualization overhead. Also, most VM placement heuristics disregard the topology of virtual clusters, resource usage patterns of each application, and competing workloads. In this paper, we study the case of scientific applications in virtual clusters by analyzing how different VM profiles and placements can affect observed performance of a parallel application that uses distributed memory. We propose a description for the characteristics of a virtual cluster through the placement of virtual cores, analyze different configurations with a software for systematic execution of virtual clusters, and explain the observed performance in terms of virtual cluster features and resource usage patterns of the application. Our analysis shows that other factors besides the number of cores can have significant effect on performance, such as virtual core mappings and VM spreading. We discuss how our methodology can be extended toward developing performance models, which are aware of resource contention and virtual cluster topology. Copyright © 2014 John Wiley & Sons, Ltd.

show abstract

Section: Related Studiesmentioning

confidence: 99%

An analysis of definition and placement of virtual machines for high performance applications on Clouds

Evoy

Mury

Schulze

2014

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Kijsipongse et al[115] propose an architecture which dynamically provisions VMs from a remote cloud according to one of the two provisioning policies introduced by the authors and adds the VM as part of a local cluster (management) system. A test-bed is setup where Eucalyptus cloud is used, which can host a maximum of six VMs on the cloud.…”

mentioning

confidence: 99%

Performance-Oriented VM Provisioning in Clouds

Shoaib¹

2021

Preprint

View full text Add to dashboard Cite

Managing applications on the cloud requires extensive decision making on the part of the Application Provider (AP). When an application faces changing workload, the services of the application are either scaled up or down in response. The services run on Virtual Machines (VM) or container instances. APs decide on how the application scales through VM provisioning and the placement of the services on the VMs. Various drivers guide this decision making. Application performance and cost are two such drivers. This thesis answers the question of how APs can meet the performance constraints of their applications while minimizing the cost of the running VMs. Two versions of the problem are presented. The first version expects to meet mean response time constraints given a deployment configuration through the replication of VMs and addition of virtual processors. The presented solution is based on layered bottlenecks. A case study shows the solution meets response time constraints and uses fewer resources in comparison to a simple utilization based approach. The second version adds the minimization of cost as an objective, where VM-types having different cost rates are used. This problem does not require a deployment configuration and provides a complete solution, where resources can be added and removed. A novel solution based on the layered bottleneck strength value with genetic algorithm has been presented. For the case study, a decision maker is implemented for a web application. The proposed solution is compared with three algorithms, all of which run within the decision maker. The results from the case study show that the proposed solution provides shorter runtime than the exhaustive search, and is able to meet response time constraints with near optimal minimization of cost. The solution also results in better cost than a plain genetic algorithm and random search, at the expense of slightly longer runtime.

show abstract

TARNN: Task‐aware autonomic resource management using neural networks in cloud environment

Sujaudeen¹,

Mirnalinee²

2019

Concurrency and Computation

View full text Add to dashboard Cite

Summary Resource management is one of the major issue in cloud computing for IaaS. Among several resource management problems allocation, provisioning, and requirement mapping directly affects the performance of cloud. Resource allocation signifies assignment of available resources to different workloads in an economically optimal manner. Precise and accurate allocation is required to maximize the usage of resources. Current method of task allocation do not take previously acquired knowledge, type of the tasks, and the QoS parameters altogether into account in the allocation phase, and it has not been trained for different set of tasks. Furthermore, the self‐optimization of the autonomous system fails to address the task type and identify the relationship between tasks and the resource demands along with its requirements. Important aspects like task management and resource utilization are the primary factors to consider for such a characteristic. This paper will present a novel autonomic resource management framework named task‐aware autonomic resource allocation strategy using neural networks (TARNN), which aims to use knowledge about the behavior of the task over an extended period of time and use this knowledge to allocate resources when a similar task is submitted in future by the user. To effectively do the allocation, a neural network–based approach is adopted to classify the tasks appropriately based on the task parameters, task type, and QoS parameters and allocate the resource optimally for a new task autonomously, without the intervention of the cloud provider. Moreover, to identify and improve the relationship of the tasks with the resources in the context of scheduling, we have proposed a novel modified Particle Swarm Optimization (m‐PSO) algorithm to schedule the tasks to resources based on resource demands. In TARNN, we have separated the collected synthetic dataset into 60‐40 ratio for training and testing purposes. We found that the neural network–based approach provides almost 80% accurate classification w.r.t. task type and QoS parameters. We have also compared our results with support vector machine (SVM) and got 69% accuracy. Since the tasks are classified appropriately, the occurrence of resource reconfiguration and VM migration is drastically reduced. Hence, our system provides better allocation of resources and schedules the tasks appropriately to the resources, thereby improving the performance of the cloud.

show abstract

Autonomic resource provisioning in rocks clusters using Eucalyptus cloud computing

Cited by 9 publications

References 13 publications

An analysis of definition and placement of virtual machines for high performance applications on Clouds

An analysis of definition and placement of virtual machines for high performance applications on Clouds

Performance-Oriented VM Provisioning in Clouds

TARNN: Task‐aware autonomic resource management using neural networks in cloud environment

Contact Info

Product

Resources

About