Managing and protecting all enterprise data


Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Midrange arrays pack in more features

Most storage array vendors have moved many features once considered "enterprise only"--such as replication and mirroring, snapshotting and deduplication--into their midrange array systems, a trend that reflects a shift in how storage needs are being dealt with; namely, storage managers are building out middle and lower tiers of storage before investing heavily in high-end systems.

Users seem to be saying, "Anything that's not totally mission-critical, let's put it on this broader midtier and let's build it out big; let's make it scalable, make it responsive to our needs and give it all these features that we want, and still get some of the top-tier capabilities like replication and snapshots," says Andrew Reichmann, an analyst at Forrester Research based in Cambridge, MA. It makes sense, given that most storage managers must meet service-level agreements, contend with disaster recovery and address data growth, pressures that are blind to the size of the storage budget.

"I never understood why people have felt that ... big customers need more functionality than a smaller customer," says Karen Dutch, NEC's general manager of advanced storage products. "Small customers still have mission-critical data; they can't afford to lose their business [either] if their data is lost." NEC began shipping its D-Series line of storage arrays last May. Included in any D-Series system is the ability to intermix SATA and SAS drives, RAID support (including RAID-6), and mirroring and virtualization capabilities. A base system with these features can be purchased for as low as $15,000.

Storage managers' buying behaviors seem to reflect the growing popularity of midrange arrays. Storage magazine's spring 2007 Purchasing Intentions survey revealed that midrange arrays will make up 43% of all array purchases this year, while high-end arrays will make up 22% (see "Storage managers in control," Storage, May 2007).

With enterprise and midrange systems' functionality rivaling each other, it's tough to know when a monolithic array will be the better bet. Vendors stress that high-end systems are best for high availability. "The way the controllers are architected, the way the front and back end are architected in the arrays, it's meant to be serviceable, expandable, while applications are running and while the arrays are processing I/O," says Kyle Fitze, director of marketing for the SAN division at HP. That may or may not be the case, depending on how the system is configured for a given storage environment.

"I've seen midrange systems outperform some vendors' highest end systems by a third, using less drives and cache to do it," says Greg Schulz, founder and senior analyst at the StorageIO Group, Stillwater, MN.

--Trina MacDonald

Article 15 of 26

Dig Deeper on Primary storage devices

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.

Get More Storage

Access to all of our back issues View All