Global file search is harder than you think

Everybody wants in on the rising market for global file search applications and services, because it's hard to do but incredibly useful if you can master it--high risk, high reward.

Image: vaniatos, Getty Images/iStockphoto

Enterprise search software is becoming very high tech. It didn't used to be that way, and now the people implementing it need to consider many new factors to prevent end-user rebellion.

Once upon a time, enterprise search was easy. Software ran queried text strings across all the files on your network--results were fast and accurate. Then came virtual LANs, wide-area networks, network-attached storage, storage-area networks, mobile devices, encryption, virtualized servers, containers, cloud storage, and unstructured data.

SEE: Software usage policy (Tech Pro Research)

"Right now, the vast majority of storage growth is based on unstructured data," said analyst Steven Hill of 451 Research. That growth too often goes into the backup oubliette, he said, not entirely joking. "Companies can't afford to do that anymore when you're talking about storage growth into the exabytes."

Another modern search challenge is finding and weeding out personally identifiable information, which could reside anywhere from a public cloud all the way down to your tape library, he noted. It's a big problem to solve, so everyone from name-brand giants--Amazon Web Services, Citrix, Dell/EMC, Google, HP, IBM, Microsoft, Oracle--through a gamut of specialists, startups, and also-rans--are trying to get traction as the top players.

How to select a global file search product

Hill offered three tips for selecting the right global file search product.

  1. First, know your use case: There should be more than one, or else your investment may not be worth the cost.
  2. Second, make sure it will work with all your disparate applications, data types, and infrastructure.
  3. Third, see if the results you'll get from the software will provide more insight than just telling you what information is where.

That third point is salient. "I think the next big challenge we're looking at in the industry is data identification," Hill added. "It's going to be a recursive process. But we need to start thinking about it now. We should have been thinking about it 20 years ago."

Sometimes the application that best meets your needs will be installed at your data center, but if you're looking for an application that uses machine learning, then it may be best to seek a cloud product. That's because a major cloud provider will generally have better hardware support than you could do on your own, Hill observed.

SEE: Cloud v. data center decision (ZDNet special report) | Download the report as a PDF (TechRepublic)

How a startup is trying to address these issues

One company trying to address these problems is the startup Cloudtenna, where CEO Aaron Ganek emphasized an investment from Citrix to illustrate his company's chances for success."What we found, when people are solving a global-file-search-in-the-enterprise problem, is first of all until recently this problem has been ignored," Ganek said. "Employees use six different tools to get their work done."

Ganek said a modern approach to contextualizing global search results is to consider the behavior and trends of files and users, not just Google-style link-back ratings. And yet there are still problems no company can truly solve right now, such as making a global search product that seamlessly integrates with cloud applications such as Salesforce or Slack. Not even email search is integrated well enough into global search products industry-wide, he acknowledged. Moreover, users are coming from the Google experience and want their results within one second or they think something's broken, he noted.

Nobody's living happily ever, but the right shopping and implementation process can help bring your company's search tools out of the realm of fairy tales.

Also see