News Stay informed about the latest enterprise technology news and product updates.

Startup Kazeon promises deeper file management

Kazeon Systems has plans to ease data classification with a technology that finds and classifies unstructured data based on the content in the files.

Fresh out of stealth mode, startup Kazeon Systems is hoping to make waves in the data classification space with the promise of search and retrieval features that not only find data, but rank it, move it and protect it.

The Mountain View, Calif.-based Kazeon will focus on unstructured data, which it defines as non-transactional data such as shared files and e-mails. The company said it will be targeting enterprises facing increased compliance regulations. "Unstructured data sitting inside e-mails and files has become more important because this is information that regulators are now looking for," said Sudhakar Muddu, Kazeon CEO and co-founder.

Muddu added that the problem with managing unstructured data is that many companies can't find it with traditional text searches, because the data formats are application-dependent, and that just basing searches on metadata doesn't go deep enough. Muddu described the Kazeon technology as "content-aware," meaning that it classifies data based on the content in the files, and not merely on the metadata.

Related articles

Crash Course: Compliance

The state of ILM

How to select an SRM suite

Avoid costly errors with e-mail archival

Data classification itself is nothing new. Major vendors like IBM Corp., Hewlett-Packard and Veritas Software Corp. have software that can classify information based on metadata. This provides data about the data, such as when it was created and how big the file is. But by classifying files based on the information inside them, Kazeon plans to provide faster file retrieval for companies under pressure to deliver data to compliance regulators.

"This technology is kind of like opening up a box and looking through the contents, as opposed to just looking at what's written outside the box," said Brad O'Neill, senior analyst, the Taneja Group.

The Kazeon technology works by sending out file crawlers throughout a network to find data in file systems and pulling all information about the file into the Kazeon system. There the files are cracked open and text is extracted and put in a repository where the file can be searched, flagged, copied and moved based on policies set by the user.

Enterprise content management (ECM) systems provided by such companies as EMC/Documentum and FileNet Corp. provide data classification similar to Kazeon. But according to Troy Toman, vice president of marketing at Kazeon, "ECM products present new tools that change the way existing applications interact with files, whereas with Kazeon that interaction does not change."

O'Neill describes the "content-aware" technology of Kazeon as a key component of information lifecycle management (ILM). "Without this kind of advanced data classification, you can't have ILM."

However he mentioned that Kazeon's technology is an enhancement to existing environments and predicts it will eventually be merged with other products. "It's a technology that screams for integration," he said.

Other startups with technologies comparable to Kazeon are Abrevity Inc., Arkivio Inc. and StoredIQ.

Kazeon's first product is currently in testing and is expected to be available in mid-2005. The company has no customers it can reference at this point. Kazeon CEO and co-founder, Sudhakar Muddu, founded SAN technology company Sanera Systems, which was sold to McData in 2003. The Kazeon management team also includes several executives who hail from Cisco Systems Inc., Sun Microsystems Inc. and Veritas.

Dig Deeper on Data storage compliance and regulations

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.