International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395 -0056
Volume: 04 Issue: 03 | Mar -2017
p-ISSN: 2395-0072
www.irjet.net
HPC Cloud Burst Using Docker Khandave Devendra , Kumbhakarn Swati, Kulkarni Shruti, Janbandhu Shreejeet Pune Institute of Computer Technology, Department of Computer Technology, Pune University Pune,India ----------------------------------------------------------------------------------------------------------------------------- -------------------------------Abstract - Cloud Bursting is an application model in which an application runs in a private cloud or data center and burst into a public cloud when the demand for computing capacity spikes. High Performance Computing (HPC) application require system with environments for maximum use of limited resources to facilitate efficient computations. However, these systems are faced with a large trade-off between efficient resource allocation and minimum execution times for the applications executing on them. Also, deploying applications in newer environments is exciting. To alleviate this challenge,container-based systems are recently being deployed to reduce trade-offs. Here we investigate container-based technology as an efficient virtualization technology for running High performance scientific applications. We select docker a a container based technology. Docker is a tool designed to make it easier to create,deploy,and run applications by using containers. Containers allow a developer to package up an application by using containers. Containers allow a developer to package up an application with all of the parts it needs.In this project, we are trying to deploy the container(docker)consisting web applications from private to public cloud securely under the situation when overload occurs.
each virtual machine (VM) created. This approach requires memory resources and slows down overall execution times of applications. Containers on the other hand, do not require Guest OS thus are more light-weight compared to hypervisor-based virtualization technologies.
KeyWords Docker, Storage,Virtualization
2.1Docker Container VS virtual machine
Distributed
System,
Using Docker container-based systems, we demonstrate that the light-weight feature of container-based virtualization compared to hypervisor-based virtualization reduces the overall execution times of HPC scientific applications due to approximately zero start-up time when launching containers. . We also demonstrate that even though the most utilized resource in the Docker container-based system is main memory (RAM), Docker manages memory resources efficiently hence creating a stable environment for HPC applications. We are trying to reduce the level of effort and time required to deploy the applications. Docker containerization environment coupled with automatic configuration and deployment modules allow quickly-deployable ,easily reconfigurable solutions. 2. RELATED WORK
Security,
1. Container has less Overhead- At best, you could run may be fifty VMs on the top of the single physical host and you need a powerful server host to get to even that number. But because container have less overhead you could run hundred on the single host.
1.INTRODUCTION In these recent years, virtualization technologies have been adopted to support efficient scientific computations and high performance applications. Correspondingly, there have been diverse Cloud Management Platforms (CMP) which provision and manage various computing resources. At the infrastructural level, platforms like OpenStack are mostly used to provision and manage both private and public cloud platforms with their processor, storage, and network resources. These aforementioned middleware systems developed on the basis of hypervisor (HPV) virtualization technology however, require the installation of Guest Operating Systems (Guest OS) for
Š 2017, IRJET
|
Impact Factor value: 5.181
2.Containers are easy to work with-One of the coolest thing of docker container is the way you can pull and run a container image in a few.You could also download and set up a VM image from the internet but that process is not stream lined. 3.Containers are more standardized-There are multiple VM platfroms each with its own way of doing things.Being an expert VMware does not necessarly qualify you to work with KVM but with conatiners ,dockers dominance and open container initiative have help to standardlized the entire container stack. 4.Containers are more open-Some VM platform such as KVM are open source but most are commercial product,
|
ISO 9001:2008 Certified Journal
|
Page 1378