Virtual Server PK Legacy Storage schema

Source: Internet
Author: User
Keywords Server storage architecture can virtualization disk

Wikibon founder and http://www.aliyun.com/zixun/aggregation/5343.html "> Chief Technology officer David Floyer said The biggest problem with enterprise in the 2011 is the lack of a modern storage architecture for deploying virtual servers. The problem is like trying to change old classic cars into modern SUVs. Now that businesses are aware of the benefits of virtualization for cost control, energy conservation, and integrated management, it's not cheap to move to virtualized servers, so many organizations are trying to build new virtual environments with existing old storage infrastructures.

The

Auto Compact configuration

preconfigured is a bit outdated and is a less efficient way to fuse traditional storage architectures and virtualized servers. Administrators may provide 1TB of capacity to the server, and then end up using only 100GB, resulting in 900GB of capacity being locked out and nowhere else. In contrast, an automated compact configuration can provide all the necessary storage for a virtual server, and these storage arrays are only physically configured when data is written. This can maintain very high storage utilization while providing virtual capacity to the server as needed.

"On average, customers use only about 20% of the traditional storage," says Hewlett-Packard's Nunes, "while the server thinks it has 1TB storage, for example, the storage array allocates disk capacity only when you write data, if you write to 200GB, Then your array will only have 200GB for the application. This is a bit like VMware's CPU utilization, except that it says storage utilization. A streamlined configuration will allow your storage utilization to reach 70% or 80%.

using virtualization in storage heavy environments eliminates the biggest cost factor in new virtualization projects. There is no need to spend on unnecessary storage capacity, less time to recover costs, and higher ROI. In addition, operating expenses also declined because only a smaller number of devices were needed. If you do not have an automatic compact configuration and the preconfigured mode is still accepted, you may have the opposite result. The accessibility features of the

Auto-compact configuration are thin-recycle (thin reclamation), sometimes referred to as compact persistence (thin persistence), which involves storage capacity that is unnecessarily locked. In normal configuration mode, the administrator starts the virtual server one day, configures the storage capacity for it, and then releases the server over time. However, there is no easy way to recycle the storage that is assigned to the server, and a thin collection provides an easy way to send the deleted virtual machine capacity back to the unassigned storage pool.

HP estimates that a streamlined configuration combined with a streamlined recovery will reduce the total capacity usage by at least 50% compared to traditional configurations. "If you deploy HP's 3Par series, we can guarantee that we can replace the original storage system with just half the capacity of the original storage system," Nunes said.

Performance factors: Virtual servers and legacy storage schemas

Virtualization advocates typically advocate a physical server consolidation rate of 10:1. 10 old servers can be merged into a server running a management program. Advantages include reduction of management complexity, energy consumption, etc.

As we all know, old single application servers often do not take full advantage of their CPUs, and now this problem is through virtualThe simulation is well solved. However, now that the CPU is fully utilized, memory becomes the next bottleneck problem, resulting in increased paging to disk. Virtual machine performance is limited by storage I/O performance. The

has a high-speed storage array, and faster disk response provides lower latency and higher overall performance. This alleviates bottlenecks and enables a single physical server to install more virtual machines, no longer 10 virtual machines, and administrators can install 15 or 20 virtual machines, which is clearly evident here.

A key factor for this level of disk I/O is wide stripe. The wide strip takes a given VM volume and spreads it to each disk drive in the array. At the same time, the volume is also wide striped across each controller in the array. This kind of stripe method directly affects the maximum continuous throughput.

Management

Automatic thin configuration helps to better manage the unpredictable workload of change compared to traditional storage for predictable workloads. Changing traditional storage configurations can cause downtime, impact performance, and require more labor. Without the support of modern storage architectures, these changes, including configuring or allocating capacity at different tiers of storage, while maintaining high service levels, are challenging.

"The core of our management technology is automated management," says HP's Nunes, "where virtualization and automation within the disk array can handle the workload of virtual servers from all of these links." It can be configured, layered, and load-balanced on its own, without requiring storage administrators to manually manage all of this, which will save administrators a lot of time.

In summary, for a virtual server, the storage architecture designed for virtualized environments is definitely superior to the storage architecture that existed before the smartphone or World Wide Web. The converged storage architecture enables these forward-looking advantages and helps ensure that modern enterprises gain the benefits they want when upgrading storage deployments.

(editor: Heritage)

Related Article

Contact Us

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.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.