This article can also be found in the Premium Editorial Download "Storage magazine: Adding low-cost tiers to conserve storage costs."
Download it now to read this article plus other related content.
In the 1960s, IBM changed the way business was done. It made billions because of it. In 1989, EMC changed the way storage was built, sold and supported--at the expense of IBM--and it made billions because of it. In 1992, Sun Microsystems owned every aspect of the Unix world, and then came little Network Appliance, changing the way we did NFS file serving forever--and it made billions.
Who are the next dragon slayers in storage?
First, it is important to have the right mindset when thinking such ethereal thoughts.
Rule No. 1: Having the best stuff is nice, but low on the totem pole of relevance. Having the ability to spot a weakness and attack it violently is more important. You don't think IBM or Sun could make better stuff than little tiny EMC or NetApp? IBM and Sun took their eye off a ball they didn't see as important, or they thought they owned.
Rule No. 2: Just because something worked for EMC and NetApp back in the day, it doesn't mean it will work today. Different times require different thinking.
Some of the companies I think could be the next giant killers are:
Compellent. Phil Soran, former Xiotech founder, decided to try to do the storage impossible--build an array that a monkey could operate, that could grow or shrink on the fly, that would always optimize itself for performance based on user's rules and could be completely fault tolerant and heal itself. In short, Phil decided to build the highest functioning
EqualLogic and Intransa. These iSCSI array guys are slugging it out against incumbents, and winning with a new way of doing things. Both use a self-healing, self-growing methodology to create idiot-proof, high-function storage. Like Nexsan, these guys don't need a giant services arm because if something breaks, nothing bad happens--the systems heal themselves. Users find they can manage more storage per administrator vs. traditional approaches. EqualLogic went so far as to give away replication software with its boxes--a move that threatens huge profits from the big guys. The advent of the iSCSI standard makes the cost of connection to these systems non-existent--an inflection point in technology that was necessary to make people look at something new.
Nexsan. This little player with the big heat has sold an absolute ton of ATA-based disk arrays to huge customers. It sold to the Fortune 10 when the big guys said, "They will never succeed. They have no service. They have no data center reliability." Who cares if you don't have a billion service people if your array is fault tolerant, and is so cheap that customers buy two or three and mirror them and still save money! I know more than one salesperson at a few of the big companies who are starting to get very worried about Nexsan.
There are plenty more examples of companies that seem to have the stuff to make a bid for supremacy of the immediate future. All of them have a few things in common: They are all nimble and capable of reacting to the needs of the customer faster and more effectively than the big guys. If EMC isn't careful, it's going to get out-EMC'd by some new three-letter acronym.
This was first published in August 2004