Inside IT Storage

Seagate Enterprise Inside IT Storage

To auto-tier or not when managing storage data

With today’s complexity of IT data management, and the need for ever-increasing efficiency and cost reduction, tiering your data based on its use is essential. What that means is that the most frequently accessed information stays on the higher-performance Solid State Drives (SSDs) and 15K-rpm Hard Disk Drives (HDDs) and less frequently used data can move to lower-cost nearline 7,200-rpm HDDs.

This movement also tends to reflect different levels of capacity. As the data becomes less-frequently accessed, the nearline drives handle and store this bulk data naturally through their larger capacities. For a typical business that sells a product or service to a customer, order processing could be handled on the highest-performing tier. But then after a given period of time, say three to six months, those now older order records can be moved to the nearline drives. For some businesses, using a tiered system that uses three levels of storage makes more sense than two tiers. A tiered system built around Seagate drives could mean use of Pulsar SSDs at the top tier, then 15K Cheetah or Savvio SAS drives in the second tier, and finally 7,200-rpm Constellation drives.

Tiered systems that automatically move information based on their use have emerged, and these software systems are available from EMC, HP, Dell, IBM and others to use with their own servers. Auto-tiering looks at the data and then moves it based on use patterns, enabling the IT manager to focus on other areas in the data center.

But even with auto-tiering’s convenience, some IT professionals may not yet be ready to relinquish control to a software application in an area that can impact performance and data center efficiency. Manually moving data from tiers over time does give the IT manager direct control over exactly what is moved and when. And an automatic tiered solution may not match the needs of all businesses, especially if it is a cyclical or unpredictable one where hot data that becomes cold may then become hot again.

Automatic tiering of data is a great option — but can it be made to fit the needs of all businesses over time? Or even if it could, will their always be a group of IT professionals who will want to more closely evaluate their data, and tier manually? What do you think – would you be willing to let the system automate your data tiering?

3 Comments

Post a Comment

Your email is never shared.

* Required fields

* Seagate will review all blog submissions and determine, in its sole discretion, whether such submissions will be posted for broader viewing. No blog comment will be considered for posting if deemed potentially damaging to Seagate's reputation or insufficiently aligned with the relevant blog topic. Without in any way limiting the foregoing, no submissions will be posted that contain: confidential company information; profanity; racial slurs; gratuitous references to sex, substance use, or violence; or statements that are in any way contrary to the letter or spirit of Seagate's Code of Business Conduct and Ethics.