Honeycomb full of holes?

by Robin Harris on Tuesday, 4 January, 2005

Sun is doing some reasonably smart pre-announcing of what they hope will be the Next Big Thing from their beleaguered storage group: Honeycomb.

This system promises to: “search and retrieve files at a quick clip. The Honeycomb boxes will run specialized indexing software to pull this off. . . .” Very cool, conceptually.

So it is NAS on steroids, with Google desktop wrapped in. From the 100,000 foot level this sounds great. But will customers go for it in a big way?

No, with one exception. If you have millions of files that you need to search often enough (think the next Microsoft anti-trust case) that a distributed index and search engine would make a difference, then this could be for you. But this is a niche.

The fundamental problem with Honeycomb is that there is no compelling reason to have that intelligence in the storage box rather than the network. Functionality that optimizes its price/performance by being on the network usually migrates there, which is clearly the case here.

Why? First, there is a large installed base of NAS boxes already, with a lot of data that people are already interested in accessing, and that is readily available on the network. Why not index this data as well?

Also, customers don’t like lock-in. Once you’ve bought a bunch of these babies, what are you going to do with them in five years when it is time to upgrade?

There is a great opportunity for network-based NAS data management out there. Sadly, Sun is missing this opportunity with a solution that is both limited and proprietary.

Honeycomb is heavy on holes and light on substance.

{ 0 comments… add one now }

Leave a Comment

Previous post:

Next post: