Thin provisioning technology solves the problem of wasted disk space

The advantages of cost savings and time resources make the thin provisioning technology increasingly popular with users. Advanced savings concepts also reduce the cost of communication lines and the complexity of storage architectures for remote mirroring.

In the past, the way the server stores data in a direct-attached hard drive is to assume that the hard drive does not change (that is, the capacity is fixed) to create an operating system and applications, which is a major technical problem left over. Although most servers and applications have stored files in variable capacity, logical unit (LU)-enabled array subsystems over the past decade, these logical units are still treated as fixed-size units or physical units.

The handling of operating systems and applications is difficult to change in the short term. This brings a huge hidden burden to the corporate IT department that needs to support a large number of servers. Thin provisioning can reduce the cost pressure of traditional storage configuration solutions.

Savings in Thin Provisioning

Thin provisioning is a storage management feature that provides a very large virtual drive for the operating system, albeit with much smaller actual capacity. For example, the storage system actually allocates only a few gigabytes of space, but provides a 2TB LU for the operating system. The application "thinks" that the LU is very large, however, in fact most of the space is virtual. As more and more data is written by the application, the actual storage space can be expanded in time.

A significant advantage of this technology is the ability to automatically expand the allocation volume without manual expansion. Another big advantage that is less likely to be discovered is that it enables users to achieve nearly 100% storage space utilization because all of the actual allocated space corresponds to the actual data written by the application.

When the thin provisioning technology is combined with remote mirroring technology, it will show another big advantage. In this case, in addition to saving a lot of bandwidth needed to build the image, it also saves disk space on both sides of the image. In the initial process of Block-based remote mirroring, the entire volume is transmitted over the WAN. A thin provisioned volume allows you to transfer only the actual used space without having to transfer the entire volume. This greatly reduces the time required to build an image, as well as enabling low-cost communication lines for remote mirroring.

Saving more than just money

Calculating how much money can be saved by thin provisioning technology, the most direct way is to measure the increase in storage capacity utilization. We only need to look at the current capacity of a particular application, then load the application into a thin-provisioned volume and calculate the actual occupied capacity, then add up the annual savings to get the total savings. Typically, thin provisioning technology increases storage capacity utilization from 60% to 80%, which saves 20% in storage costs per year. In fact, thin provisioning technology can achieve even more savings.

- IT Costs and Efforts to Extend Logic Units

For most applications, no matter how much storage space is pre-allocated, space will definitely not be enough at some point. Generally speaking, the volume of the volume needs to be expanded every two to three years. Depending on the capabilities of the application, operating system, and storage array, expanding volume capacity typically requires hours of hard work by IT operators. Every storage volume is expanded every three years. For systems with more than 150 volumes, this means an average weekly expansion.

- Costs associated with downtime due to extended volumes

In general, extended volumes require downtime. While some applications, such as Exchange, can support instant volume expansion, new storage capacity can only be used after Exchange is restarted. Depending on the application, the restart requirements are not the same, but usually involve cost factors, such as reduced production due to downtime, overtime on weekends or overtime work overtime. In a system with hundreds of volumes, you might work for it every week.

Thin provisioning technology avoids the downtime caused by this.

- Costs related to electricity and cooling

Storage arrays are large consumers and require a large amount of cooling resources. If the thin provisioning technology can increase the storage capacity utilization by 20% to 30%, Then consume as much power to store more data. When calculating the return on investment, it is clear that this issue should be considered.

Thin provisioning technology is gaining popularity among users because of its significant cost and time resources. In addition, the technology is expected to further reduce the complexity of current storage infrastructure.

However, as with any new technology, there are many different implementations of thin provisioning technology. New challenges and problems will be encountered in the implementation process, and we must consider them carefully. Because not all applications are suitable for thin provisioning technology, and as technology continues to evolve, users will encounter more new problems.