Method and system for providing virtual desktop service using cache server

Patent metadata

Field	Value
Publication number	US-10037223-B2
Application number	US-201615086234-A
Country	US
Kind code	B2
Filing date	Mar 31, 2016
Priority date	Apr 1, 2015
Publication date	Jul 31, 2018
Grant date	Jul 31, 2018

Abstract

Provided herein is a system and method for providing a virtual desktop service using a cache server, the system including a host server configured to provide the virtual desktop service to a user's device using a virtual machine, a distributed file system configured to store data for the virtual machine, and a cache server configured to, when a request to process data reading is received from the virtual machine inside the host server, pre-read a data segment for the requested process of data reading from the distributed file system and store the pre-read data segment in a cache, and perform the process of data reading using the cache.

First claim

What is claimed is:

1. A system for providing a virtual desktop service, the system comprising:
a host server configured to provide the virtual desktop service to a user's device using a virtual machine;
a distributed file system configured to store data for the virtual machine; and
a cache server configured to, when a request to process data reading is received from the virtual machine inside the host server, pre-read a data segment for the requested process of data reading from the distributed file system and store the pre-read data segment in a cache, and perform the process of data reading using the cache;
wherein the cache server obtains information on the data segment from read distance maps each of which includes a plurality of blocks arranged in a predetermined format, each block corresponding to the data for the virtual machine and including information on a next block in the predetermined data segment to be read after a current block is read; and
wherein each of the blocks provided in the read distance maps comprises:
i) first information indicating reusability of each block, ii) second information indicating, when each block is a block constituting the data segment, whether the data segment is based on temporal locality or spatial locality, iii) third information indicating a value expressed as the data segment based on the spatial locality where the special locality represents the offset value to the next block read, and iv) fourth information indicating a value expressed as the data segment based on the temporal locality.

2. The system of claim 1 ,
wherein the data segment is at least one of a booting segment and an application segment, and
the cache server determines whether the request to process data reading received from the virtual machine is for a booting or for an application, and uses the read distance map corresponding to a result of the determination.

3. A method for providing a virtual desktop service, the method comprising:
prefetching, when a request to boot a virtual machine is obtained, a booting segment to a cache with reference to read distance maps for booting, each of which includes a plurality of blocks arranged in a predetermined format, each block corresponding to data for the virtual machine and including information on a next block in the predetermined data segment to be read after a current block is read;
analyzing, when a request to process reading of predetermined data is received from the virtual machine after a booting of the virtual machine is completed, whether or not the predetermined data requested is for starting an application;
checking, when it is determined as a result of the analyzing that the predetermined data requested is for starting an application, in which one of the read distance maps for applications the predetermined data is provided as a block;
prefetching, when it is determined that the predetermined data is provided as a block in one of the read distance maps for applications, an application data segment corresponding to the predetermined data to a cache using the read distance map in which the block is provided; and
wherein each of the blocks provided in the read distance maps comprises:
i) first information indicating reusability of each block, ii) second information indicating, when each block is a block constituting the data segment, whether the data segment is based on temporal locality or spatial locality, iii) third information indicating a value expressed as the data segment based on the spatial locality where the special locality represents the offset value to the next block read, and iv) fourth information indicating a value expressed as the data segment based on the temporal locality.

4. The method of claim 3 ,
further comprising, when the booting being an initial booting of the virtual machine, creating a booting segment that refers to data being used at the initial booting in one of empty read distance maps.

5. The method of claim 3 ,
further comprising, when it is determined that the predetermined data requested is not provided as a block in one of the read distance maps for applications, creating an application data segment for the predetermined data in one of empty read distance maps.

6. The method of claim 3 ,
wherein a data segment for re-use that different data segments use in common is prefetched before receiving the request to process reading of the predetermined data from the virtual machine after the booting of the virtual machine is completed.

7. The method of claim 3 ,
wherein an age value of the blocks corresponding to the data segment prefetched to the cache is increased.

8. The method of claim 7 ,
wherein the age values of the blocks of the read distance maps for booting and of the read distance maps for applications are checked at every predetermined period of time, and any block having an age value of 0 is removed from the read distance maps for booting or the read distance maps for applications.

Assignees

• Electronics & Telecommunications Res Inst

Inventors

• Park Jeong Sook
• Kim Young Kyun

Classifications

• G06F17/30238

  Physics · mapped topic

• G06F2009/45583

  Memory management, e.g. access or allocation · CPC title

• G06F17/30203

  Physics · mapped topic

• G06F2009/45575

  Starting, stopping, suspending or resuming virtual machine instances · CPC title

• G06F9/45558 Primary

  Hypervisor-specific management and integration aspects · CPC title