Close this search box.

Hybrid Storage Makes Storage Tiering in the Virtualized World Irrelevant; Interview with Tegile Systems VP Rob Commins Part III

In the fast-paced, ever-changing world of virtualization, the ability for a storage array to deliver performance at exactly the right time is essential. Unfortunately, most tiered storage systems are poorly equipped to respond to these new dynamics. This is where hybrid storage arrays come into play. In this third installment of my interview series with Rob Commins, VP of Marketing at Tegile Systems, we discuss the practical applications of storage and data movement in a virtualized world, how storage tiering falls short of consumer requirements and why Tegile’s hybrid storage is so well-equipped to meet them.

Jerome: What are the key technologies that differentiate Tegile Systems from its competitors?

Rob: The market has progressed in terms of addressing this performance density issue. If you rewind 5, maybe even 10 years, some vendors broke the mold of having RAID groups inside of an array. Maybe you have an array that has 100 disk drives in it. Then you break the array into 10 RAID groups with 10 drives each. With a 15,000 RPM drive delivering about 200 IOPS a drive, an individual volume could only get 2,000 IOPS.

When you implement this notion of wide striping, which would actually create one very large pool of disks, those 100 disks become virtualized and individual volumes get striped across all 100 disks. Then you can get an individual volume that can get 20,000 IOPS out of it. Since you’re taking a razor thin slice of all of those spindles, you usually run into an issue where you have got gobs and gobs of more capacity than you need and your power space cooling costs become irrational.

The second step is then people took, say, 10 of those 100 drives, yanked them out, and put in 10 solid state drives (SSDs). What happens then is the people started using some tiering software that would sweep the array for what I like to call, “IO density.” You can find the hotspots of data and migrate data up and down those tiers – maybe a flash tier, a 15,000 RPM tier, and a 7,200 RPM tier – moving the data to where it belongs economically and performance-wise.

The issue we found is this: In the virtualized world we live in, users and their data are moving around a lot faster than what the tiering software can handle. The default sweep time for that IO density on the tiering software is usually between one hour and three hours.

Some of them sweep and move data at a very quiet time of 2:00 a.m. But if you have a virtual desktop environment that is hitting a boot storm at 8:00 a.m., you do not have three hours let alone another whole day to wait for that system to figure out, “I better promote that data to flash.”  Your boot storm is over in an hour and a half.

Reaching back to the retail example, you may have a database that is getting pulled at the end of the quarter. If you have somebody doing a big pull on a database, it is probably done in a matter of minutes. That sort of data pull is long gone by the time this tiering software has figured it out.

That is why hybrid is so neat. It uses caching algorithms. Data is moved right now and kept in a large pool of data reduced cache. Tegile can address that timeliness of performance issue that those other two architectures cannot.

Are there other environments you want to comment on?

Rob: We happen to do a lot of Microsoft Exchange and SharePoint deployments. At the end of the day they are just ones and zeroes to us. We sell to higher education, financial services and healthcare. We are very horizontal that way, although we do have a nice piece of our business in the service provider market. They are able to take those cost reductions from deduplication and compression and meet really nice performance SLAs with hybrid while passing that savings right into their pricing for the end users.

: What is Tegile Systems’ largest installation?

Rob: Tegile has many customers that have passed the 10 system mark. Tegile also has several customers that have made full migrations from legacy tiered storage systems to Tegile’s Hybrid Arrays on their entire data center floor.

In Part I of this interview series, we talked about how sub-millisecond response times are the new gold standards in storage system performance.
In Part II of this interview series, we examined what the tipping point is when going from hybrid to flash.
In Part IV we will discuss the benefits that ZFS brings to a relatively new storage player like Tegile.


Click Here to Signup for the DCIG Newsletter!


DCIG Newsletter Signup

Thank you for your interest in DCIG research and analysis.

Please sign up for the free DCIG Newsletter to have new analysis delivered to your inbox each week.