Browse Prior Art Database

Method of accelerating virtual machine doployment, optimizing resource management and enhancing performance and security in cloud environment

IP.com Disclosure Number: IPCOM000248282D
Publication Date: 2016-Nov-15
Document File: 6 page(s) / 95K

Publishing Venue

The IP.com Prior Art Database

Abstract

There is a hypervisor whose job is to manage life cycle of virtual machines running on the platform. It uses images to deploy virtual machines with similar business workloads, as lots of other cloud platforms do. The difference is that our hypervisor will not write image data onto the disks of the virtual machine. When deploying a virtual machine, it load the image data into a memory area, and then mark the area as read only.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 6

Method of accelerating virtual machine doployment, optimizing resource management and enhancing performance and security in cloud environment

It's cool to have business servers running in the cloud rather than on an iron shelf. But it's not because they are running in the cloud, it's cool because it's quicker to deploy servers, resources are utilized more efficiently, applications are running faster, and business data are moresecure. But it's not the end of the story! Deploying a virtual machine is faster than a physical machine, however, it's still not fast enough, especially when there are dozens of virtual machines waiting for being deployed. Virtual recourses are utilized more efficient than physical separate resources, however, it's possible to consume even less, especially in a case that most data among virtual machines are common. Applications in cloud are usually running fast because it's easy to handle jobs concurrently in a cloud, however, for each individual thread it's usually running slower than in a physical server. A good cloud usually is hard to be hacked in, however, in case it is it's then business data must not be lost in the disaster. This innovation is the way to get closer to those targets.

There is a hypervisor whose job is to manage life cycle of virtual machines running on the platform. It uses images to deploy virtual machines with similar business workloads, as lots of other cloud platforms do. The difference is that our hypervisor will not write image data onto the disks of the virtual machine. When deploying a virtual machine, it load the image data into a memory area, and then mark the area as read only. There will beno modification of the area until the next time of booting the system, or the garbage collector decides to release it. All virtual machines born from the image will share the data in the memory, and only their particular data which are different from the image will be stored on their local disks, like network configuration data, etc. Benefitting from eliminating the time of copying image data onto disks, deployment becomes quite fast. The time could decline form minutes to seconds. And since only a few data will actually be written on the disks, only a small amount of disk capacity is very enough. The hypervisor could assign a small disk capacity to the virtual machine, and give it more in the future as it needs, which could save a huge amount of disk capacity.

When the virtual machine begins to work, it can read common data from the memory. Physical disk is accessed only when...