Driving Storage Growth

Enterprise Storage Forum content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.

There was a time not too long ago when a 40GB drive was the biggest single drive anyone could buy.

As with other sectors of information technology that see continual performance improvements, storage capacities are similarly on a seemingly boundless growth curve.

Once seemingly impossible storage capacities are now in full production. But is there a limit to where capacity can grow? And what effect is higher capacity storage having on the enterprise?

Is There a Limit?

“In general, there are no limits, and in fact any theoretical limits would be very difficult to estimate,” Seagate senior product marketing manager Gianna DaGiau told Enterprise Storage Forum.

Time and again, drive vendors, like their peers in other sectors of IT, have found ways to break through perceived barriers. On magnetic hard drives, which are expected to continue to be the norm for some time to come, it’s all about how many gigabits per square inch, or “areal density,” a drive vendor can jam in.

“There was a time when achieving an areal density of 100 gigabits per square inch was considered impossible,” DaGiau said. “However, Seagate found ways to achieve this density and move past it.”

Among the leading areal density innovations being trumpeted by Seagate and other drive vendors is the concept of “perpendicular recording.” Instead of the traditional longitudinal recording method where bits are arranged horizontally, perpendicular recording arranged the bits vertically across the surface of the disk. Seagate first announced its perpendicular breakthrough in 2002 (see Seagate Demonstrates Record Storage Areal Density).

Along with an innovation called Heat Assisted Magnetic Recording (HAMR) and other technological improvements, DaGiau contends that the disk drive has quite a long lifespan ahead of it, with no limitations in site.

Craig Butler, brand manager for disk storage product marketing at IBM, echoed Seagate’s optimism.

“We are not planning on anything but the current curve, where we’re getting a doubling of capacity every 18-24 months,” Butler said.

Info-Tech research analyst Curtis Gittens said vendors will roll out perpendicular recording technologies in their 3.5″ drives, and 600GB to 700GB drives will debut later this year, with 1TB not far behind.

“Given the growth of storage and the innovation drive manufacturers continue to show in both storage and tape, I anticipate that we will see the 1TB drive by late 2007-early 2008,” Gittens told ESF.

What It Means for Enterprise Storage

The increased capacity of drives is also driving increased total capacity available in storage arrays. EMC recently announced a new Symmetrix DMX-3 that could power up to 2,400 drives and scale to more than one petabyte of capacity.

Over at IBM, Butler noted that their biggest single box is the DS-8000. Behind a single controller you can put 192 TB, but the system is architected to expand to 1.2PB. Grid storage farms of multiple storage arrays will exceed the figures too.

Stuffing more drives into each array is one thing — EMC’s DMX-3 boasts a capacity of up to 2,400 drives and IBM’s off the shelf DS-800 packs in 640 drives — but when you can put more storage on each individual drive, the overall total capacity potential increases exponentially.

Higher-capacity storage is also serving to help drive down storage costs. In Butler’s estimation, drive costs are currently declining at a 33 percent compound annual rate. He expects that enterprises will be paying $5 per GB of high-end storage by the end of 2008. SATAcosts will be even lower and are expected to be in the $3.50 per GB range by 2008.

What are Enterprises Buying?

In the case of SATA, end users want the highest capacity drives, according to Butler. The industry is now shifting from 250GB or 320GB drives to 400GB to 500GB drives.

“In terms of SATA, it’s about putting the most amount of storage on the lowest cost,” Butler said. “On Fibre Channel, most want the next notch down from the highest and are seeing a balance between performance, capacity and price.”

For IBM at least, even though 300GB Fiber Channel drives are available, the most common-selling drives are currently 146GB.

Seagate’s DaGiau noted a similar trend for high-capacity drives in ultra high performance environments.

“From a drive perspective, what limits adoption of higher capacity drives is the need for a certain level of performance per gigabyte,” DaGiau said.

Applications with high performance-per-gigabyte requirements typically use lower capacity drives, and applications with low performance-per-gigabyte requirements often use the highest capacity drives available.

DaGiau said that in terms of actual capacity capabilities, areal density grows independent of the physical drive interfaces, whether its FC, SASor SATA.

Capacity, Reliability a Tradeoff?

Although there may be a performance to gigabyte relationship, there isn’t necessarily a relationship between higher capacity and reliability, according to DaGiau.

DaGiau said the reliability of a drive is determined by how it is built and designed, taking into account its mechanics such as its motors, materials and build and test tolerances.

That said, there may be a difference when it comes to commodity high-capacity drives at lower price points. According to DaGiau, the general trend is that the high capacity drives that are made today are made to fit a specific price point and application. Since high-capacity drives aren’t well-suited for round-the-clock server operation, they aren’t built to the same stringent standards as an enterprise 15,000 RPM drive, for example.

That’s not to say that high capacity storage devices are necessarily any less reliable, however. DaGiau said Seagate’s latest NL35 series 400GB and 500GB models have a mean time between failure (MTBF) rating of 1 million hours, while previous generations were under that figure.

What Storage Managers Really Need

The seemingly insatiable appetite for storage is being driven by a number of well documented factors, not the least of which is compliance with data protection and retention regulations. IBM’s Butler also noted that markets such as the nuclear, aerospace and pharmaceutical industries tend to need high volumes of storage capacity.

Info-Tech’s Gittens, however, thinks storage managers want better management tools more than they want higher-capacity disks.

“Higher capacity drives are nice, and it will allow large enterprises to store more data in a single SAN/NAS device,” Gittens said. “But at the end of the day, it will mean more management headaches for storage managers if proper ILM processes are not in place. So at the end of it, I see higher-capacity disks driving more ILM implementations.”

For more storage features, visit Enterprise Storage Forum Special Reports

Sean Michael Kerner
Sean Michael Kerner
Sean Michael Kerner is an Internet consultant, strategist, and contributor to several leading IT business web sites.

Get the Free Newsletter!

Subscribe to Cloud Insider for top news, trends, and analysis.

Latest Articles

15 Software Defined Storage Best Practices

Software Defined Storage (SDS) enables the use of commodity storage hardware. Learn 15 best practices for SDS implementation.

What is Fibre Channel over Ethernet (FCoE)?

Fibre Channel Over Ethernet (FCoE) is the encapsulation and transmission of Fibre Channel (FC) frames over enhanced Ethernet networks, combining the advantages of Ethernet...

9 Types of Computer Memory Defined (With Use Cases)

Computer memory is a term for all of the types of data storage technology that a computer may use. Learn more about the X types of computer memory.