One of the best practices to improve storage ROI is to go for automated tiering. This is a big advice especially for those financial firms which are going for the deployment of Solid-State drives in their storage environments in order to make their storage environment highly performance oriented. Automated tiering, shortly named as Auto-Tiering is a key enabling technology for storage admins who want to improve workload performance, in order to improve the user experience in a magnificent way.
In general, whenever a new business critical application is rolled out, all storage admins practice over-provisioning of storage from the available storage capacity. For attaining performance, they go for the most expensive storage media for Tier 1 storage i.e. the SSD.
By practicing Tier 1 storage for all of these applications, the storage costs will naturally increase. Sometimes, it can also occupy 20% of the total IT budget. But here the only worry will be that there is a high probability of low capacity utilization. In a recent survey conducted by Information Weekly, it was revealed that storage managers were still using only 54% of their installed storage capacity and the rest of the resources were getting simply wasted.
It is to be notified over here that storage tiering is based on the value of data and its access and retention needs, in an enterprise environment. And here most of the storage admins try to overlook other issues, in a bid to support only these two issues. Storage costs, storage underutilization, storage resources getting wasted and increase in storage costs are some of the issues which are often not adhered on a proper note.
In the beginning, storage tiers were external to storage subsystems on a traditional note. The tiers included different media like tape, disk and flash. They were also much influenced on the basis of performance, capacity and cost incurred for the storage. These tiers have been defined as Tier1, Tier2 and Tier3.
In Tier1 concept, the most expensive media like SSD is used to keep the performance of the system on par with the user choice. Tier 2 is basically a less expensive media, where the access to this tier is necessary, but takes place on an occasional note. Tier 3 is usually archive based storage level where the data is kept in cold storage in order to support future enterprise business aspects.
As the inclusion of tiering started to pick up, it started to turn into an ongoing and complex process. Traditionally, admins were pushed to assign a cost base to their storage and then define tiers accordingly by manually classifying the data. Storage tiers would then be applied to applications and the data which was a part of that app. Though, there were tools to assign data to different storage tier levels, the assignment of data to the right tiers proved complex even with the presence of these tools.
Here’s where the much needed efficiency was quenched with the help of automation. The way automated storage tiering has evolved as a right solution is truly amazing; particularly for virtualized and cloud based environments, where performance is crucial to these businesses.
With the help of auto tiering software, storage admins can flexibly optimize utilization while reducing latency. This also automates the whole process of classifying data at the application layer and helps in diverting it to the optimum tier which best suits its integrity. Thus, it helps in maximizing the storage space utilization.
The auto-tiering feature also helps in easing the management of the storage admin by classifying the data on the basis of whether it is hot or not. Under normal usage circumstances, almost 50% to 60% of data is frequently used by users within 30 days of its generation. But after the said period, the said percentage gets reduced to 10% to 20%. So, at this stage, the data needs to be pushed to the cold storage layer. Here the auto-tiering feature helps in simplifying the management of the storage admin to a great extent and helps in altering the hot and cold data in between the different storage (tier) levels.
Additionally, the user demand for data can be gauged on the basis of Input/Output per second throughput or bandwidth requirements required by/for each application. Based on these factors, further monitoring and analytics lets the admin configure the auto tiering software for specific applications needs. The software intelligence when pre-configured also controls the movement of data to lower and higher tiers when usage increases or decreases.
As per a survey carried out by IDC, approximately 60% of IT shops have implemented storage tiering in their environments to improve capacity utilization. Almost the same percentage has agreed that it has SSD deployed in its automated storage tiering architecture. This clearly suggests that auto-tiering is a must have for SSDs and highly virtualized and cloud based storage environments. This is because of the need to keep active data in the fastest tier as close to real time as possible.
Citing these benefits, most of the data storage vendors like StoneFly, Inc. are coming up with products enriched with auto tiering feature.
On a conclusion note, auto tiering should be included as an evolving storage strategy. This helps in improving the ROI and reduces TCO by focusing on the value of data and its integrity, while enhancing the management capabilities and other initiatives such as Governance, Risk and Compliance on a true note.