CNET también está disponible en español.

Ir a español

Don't show this again

Christmas Gift Guide
Tech Industry

The lure and risk of storage systems stocks

With storage start-ups popping up all over the map, finding the winners is proving tougher than ever. Philip Rueppel offers a guide to the perplexed.

Why is storage, which to most people seems like a hopelessly dull area, a lot like the sexy optics and photonics field?

Simply put, both are big, fast-growing areas with lots of new technology and lots of entrepreneurs battling it out. But with the good comes the bad: Just as the optics field became over-funded by venture capitalists and other investors, the storage market is also falling victim to its own popularity.

And the end markets are in turmoil with storage pricing falling and Web-related demand for data storage falling way short of earlier forecasts.

In past columns I have extolled the positive nature of the storage business. Nothing about that has changed. We continually create more data and information, need to store it, and rarely throw it away. Coupled with the continued trend toward digital media--if you think an MP3 file is big, wait until you see the size of a digital MPEG movie--that drives robust demand for storage systems.

But making money investing in storage stocks nowadays is proving tougher than ever.

Most of the public companies had tremendous run-ups in 1999 and 2000 only to suffer spectacular drop-offs in 2001. Yet P/E multiples remain high, and near-term business fundamentals remain cloudy.

What has changed is that a new crop of emerging private companies is chasing the same pot of gold. And fueled by significant venture capital, their ranks are swelling.

I have seen four to five times as many new storage start-ups in the past year than in the previous five. But are there too many? From my angle, the answer is no--but that depends on the segment of storage that the companies are focused on.

In general, I categorize new storage start-ups into three classes:

1. New Architectures: I label these the "EMC-killers," as companies like Cereva, 3ParData, BlueArc and YottaYotta all claim new fundamental storage architectures that are more efficient and less costly than today's alternative.

2. New Technologies: This class encompasses storage-networking hardware companies like Nishan, 3ware, and Pirus, as well as chip companies like Aristos Logic and software companies in the virtualization and storage-management segments.

3. New Business Models: They offer storage as a service, like SSPs StorageWay or ScaleEight, or managed storage like Storability.

Of the three, I think the second is the most fertile area for investors. Having followed the development of the storage industry for more than 10 years, I learned that developing a new storage system is a lot harder and takes a lot longer than most technologists are willing to admit.

Because the concept of data integrity is sacrosanct, customers don't like to experiment with partially working products. Hence, large storage system products get associated with lots of testing and lengthy development delays. Just look at the difficulty StorageTek, Hewlett-Packard, Sun Microsystems and even IBM have had competing in the RAID disk-storage segment vs. EMC.

The companies I would want to invest in don't need to redefine the storage landscape. But they do need to create new technology, which complements existing infrastructure.

Elsewhere, companies that don't have a targeted, new technology will find difficulty gaining distribution. Outside of the big systems vendors (IBM, HP, Sun) Network Appliance was the last company that could afford building a storage-focused sales force. It's just too difficult competing against EMC and NetApp's sales force. To succeed today, an OEM model makes more sense.

The best example of this is Brocade Communications Systems, a focused technology play (which I would have categorized in the second group above.) It has distribution agreements with almost all of the storage suppliers, including EMC as well as the major systems vendors.

The companies in the first category are going to have to succeed at distribution if they have any hope of long-term viability.

The third category perhaps makes for the most interesting long-term prospects, but it remains a little like video-on-demand: Despite the great potential, it's being adopted slower than originally expected. Technology issues are only part of the problem, as it is difficult to maintain adequate security and performance while outsourcing storage.

But an equal challenge in my mind is that of customer momentum. Many users I've spoken with remain unsure about giving up "control" of their data. Regardless of how much better managed or more secure the service really is, many consumers still haven't fully accepted outsourced storage.

Which leaves us the middle category.

There are lots of new technologies that will improve today's class of storage systems. Infiniband and iSCSI will become viable interconnects within the next few years; Serial ATA disk drives may change the price and performance of enterprise storage, while Storage Resource Management and Storage Virtualization software will become a requirement not only at large enterprises but in small businesses as well.

From an investment perspective, the field is crowded with a lot of money chasing a few good or differentiated ideas. That leads to high valuations and higher risk for those not close to the industry. As many of these start-ups mature, we will have a number of great public companies to invest in. But that's probably after a significant shakeout. For now, stay on the sidelines.