Keeping pace with trends in 2003

Q: Which technologies will be hot in 2003, and how should I prioritize my spending and evaluation resources?

I have always been bullish on new technologies. Lately, a number of innovative companies have come to market with new, interesting technologies. And this trend is expected to continue this year.

Many of these new products do solve real problems and have believable ROIs. The catch is that most of the innovation is being driven by start-ups, and I am finding that most enterprise-level purchasers are still leery of doing business with start-ups—even when they really need the technology.

To make matters worse, dollars are in short supply, which extends the sales cycle and makes established vendors hungry enough for business that they will compete aggressively against start-ups. This means that in all likelihood many start-ups won't be around to service what they sell.

The most successful new technologies will be secondary storage products, small-to-midrange business (SMB) solutions, as well as "add-on" technologies, which can be easily installed and removed.

I am big on the SMB marketplace because there are real problems to solve here—problems that generally can be best addressed by start-ups. Why? Because start-ups usually have the technologies that solve specific SMB issues, and the start-ups are generally hungry enough to make sure the price fits the SMB budget. As for secondary storage and add-ons, these are low-risk solutions, and if they offer enough reward, vendors of these types of products can overcome the hurdle of being industry unknowns.

Farmer's forecast

I see disk subsystems getting really, really cheap, especially those equipped with ATA disks, and even those with Fibre Channel drives. Without a doubt, disk will become an integral part of the backup system. Established backup software vendors are using disk as a "staging" medium before data is moved to tape. While this is a great first step, this doesn't take advantage of disk's random-access capability—which is clearly the cooler benefit. Disks are great at storing and retrieving block-level incremental backups, and software that truly enables disk-to-disk backup will be hot this year.

The other thing end users really need is application-specific storage software. Many applications have unique storage problems, which necessitate unique solutions. A great example is e-mail. Storage solutions for e-mail will be big this year, especially if they help address the new crop of compliance regulations.

Another place where application-specific technology is missing is in snapshot agents. Just about everyone is taking snapshots these days: file systems, volume managers, network-attached storage (NAS) appliances, storage area network (SAN) arrays, etc. The catch is that a snapshot may be useless if the application or file system isn't properly quiesced.

What would be really cool is if someone solves these application-specific problems and then integrates them into a funky disk-to-disk backup solution. Imagine an external snapshot server with all of your data preserved at any point in time. Now, I would buy that, even if it came from a start-up!

As for primary storage, I think the hottest trend this year will be iSCSI. It works and it's really inexpensive. Once centralized storage becomes cheaper than distributed storage, everyone in the SMB space will buy it. Remember: SMBs have limited budgets and want the most they can get for their money. When SMBs can implement a SAN for less than the price of DAS, they will buy the SAN solution.

Finally, there are the non-invasive, add-on technologies such as SRM software, SAN monitoring software, file-system utilities, backup-system reporting tools, etc., as well as inexpensive disk subsystems and TOEs (TCP offload engines).

All in all, get ready for a host of new technologies to emerge in 2003, which effectively address the data storage and backup issues of companies of all sizes (and budgets). It's going to be an exciting year, so stay tuned!

Jacob Farmer is the CTO of Cambridge Computer. He can be reached at jacobf@cambridgecomputer.com.

This article was originally published on February 01, 2003