Virtual Desktop storage optimization technology based on data deduplication
Source: Internet
Author: User
KeywordsDuplicate data deletion virtual desktop
In recent years, cloud computing has become a research hotspot in the field of distributed computing, which refers to the application of software and hardware in the data center to provide service on demand through the Internet, and the software and hardware resources in the data center collectively as cloud, cloud computing itself is not a new technology, but a kind of service mode This model can be used to outsource equipment installation and resource management to cloud service providers, with the characteristics of billing and scalability, through the unified deployment and centralized management of IT resources, cloud computing can optimize the utilization of resources and provide users with low-cost, efficient and reliable services.
As a typical cloud computing application, virtual Desktop is a cloud application based on server virtualization. As shown in Figure 1, the Virtual Desktop Infrastructure provides remote computer desktop services to users in the cloud; The service provider runs the operating system and application software required by the user on the data Center server, and then uses the desktop to display the views of the operating system desktop view as an image transfer to the client device. At the same time, the server handles the input from the client and updates the contents of the desktop view at any time. Users can access the Internet through any terminal, through the authorization authentication can do desktop computing. In Virtual desktop computing mode, all calculations can be placed on the data center server, the requirements for terminal equipment will be greatly reduced, similar to tablets, smartphones, handheld computers and other intelligent handheld devices such as thin terminals can meet the broader, more flexible application requirements. Virtual Desktop architecture through server virtualization, centralized management, unified configuration, making computing and data more secure, while simplifying configuration, reducing IT system costs, energy consumption and management overhead.
Figure 1 Virtual Desktop Computing architecture
The virtual desktop architecture faces the storage costs and performance challenges of server virtualization. Because dozens of or hundreds of virtual desktops are consolidated onto a single physical server, with the increase of the number of users and the increase of users ' demand, the storage demand of the server will expand rapidly; If you have 100 virtual desktops on a physical server, you need to create and maintain 100 virtual disk mirrors per GB of capacity. A total of 4TB of storage capacity is required. In addition, a physical server often needs to start and run multiple virtual machines at the same time, how to optimize the virtual machine I/O access to improve storage performance is also a problem to be solved, the use of virtual disk image has a large number of data redundancy features; This paper applies data deduplication to optimize storage space utilization Because of the data weight, the data block is repeated reference frequency is different; This article will be frequently referenced blocks of data in the data center than the traditional hard disk performance of the solid-state hard drive, and also on the server side through the local disk cache hot data to improve the performance of virtual desktops.
1 Related work
1.1 Data deduplication Technology
Data de-duplication technology is a data reduction technique based on redundancy. First of all, it divides large data objects into smaller chunks using a certain data chunking strategy, and identifies them with the encrypted hash value of the data block. Through the fingerprint of the data block, we can improve the space utilization of the storage system and the bandwidth utilization of the network system. The timing of the data de-duplication operation is different. Can be divided into online data de-duplication and offline data deduplication. In the virtual desktop architecture, data de-duplication is more advantageous because of the fact that it avoids unnecessary data transfer and write operations. Differences in granularity of data de-duplication operations can be divided into file-level, block-level, or byte-level de-duplication. Based on the features of virtual machine disk mirroring, we find that block-level online data deduplication is more suitable for the optimization of virtual desktop storage.
At present, the data de-duplication technology has been widely used in the virtual machine environment. By analyzing the effect of virtual machine disk mirroring, it is found that the data de-duplication technology can greatly reduce the virtual machine disk mirrored storage space. Nath and others find that block-level de-duplication in a virtual machine-based storage system can take full advantage of storage and network resources, Liguori and others compare and analyze the effects of duplicate data deletion on virtual machine disk mirroring of different operating systems, and test the impact of data de-duplication on virtual machine startup and read performance. Zhang and others studied the effect of data deduplication on the online migration of virtual machines. This paper focuses on how to optimize the storage utilization and I/O performance of the virtual desktop storage system using the data deduplication technology.
1.2 Solid state Drive based on flash memory
The solid-state hard disk based on flash memory is a non-volatile memory, which is mainly composed of flash memory chip and flash conversion layer. Its data access performance is between the volatile memory dram and the traditional hard drive, and the price is cheaper than DRAM. Compared with the traditional hard disk, the random read performance is as fast as the sequential read performance because there is no mechanical head motion, but the write performance is lower than the read performance due to the high cost data erasure before the write operation. In addition, solid-state hard disk based on flash memory has lower energy consumption than DRAM and traditional hard disk.
The solid-state hard disk based on Flash has a higher cost-performance ratio and energy-saving advantage than DRAM, which makes it popular in data-intensive applications. The data de-duplication technology itself is a kind of intensive application, especially for data Block index and data block metadata access operation, the amount of access data is small but very frequent. Using the advantage of random read performance of the Flash-solid-state hard disk, using it to store the data block index and metadata can greatly optimize the access performance of the storage system. Because the virtual disk image file is fragmented after the data is stored, the read performance on the traditional hard disk is poor; This paper optimizes the read performance of the virtual disk image files by using the advantage of the solid state hard disk based on flash memory on the random read performance.
The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion;
products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the
content of the page makes you feel confusing, please write us an email, we will handle the problem
within 5 days after receiving your email.
If you find any instances of plagiarism from the community, please send an email to:
info-contact@alibabacloud.com
and provide relevant evidence. A staff member will contact you within 5 working days.