Never Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.
Innovators & Influencers: Splunk Takes Proactive Approach To Web Site Complexity
Michael Baum's latest startup has a growing list of big-name customers
December 9, 2006
3 Min Read
As the Web undergoes another wave of construction, all those new teraflops and terabytes often go awry. Companies are assembling new data centers to support soaring e-commerce, search, and video traffic, and technologies such as Ajax and virtualization are making Web sites more complex--and outage-prone. Amazon.com and Wal-Mart suffered site delays and downtime on higher-than-expected Thanksgiving-week traffic. Meanwhile, systems management has become spread out among teams in the United States, India, and elsewhere.
Into the breach has stepped Splunk, a San Francisco startup whose software crawls the Web performance data generated by companies' servers and software in search of bottlenecks that cause crashes. Its approach can save companies hundreds of hours in a given week compared with scouring for problems in printed reports, says CEO Michael Baum.
Baum, a serial entrepreneur who's sold companies to IBM, Yahoo, and Reuters and was an engineer on IBM's original PC team in 1981, started venture capital-backed Splunk (the name is a twist on spelunking) two years ago to address the complexity riddling data centers, and what he says is the inadequacy of IT tools to fix it.
By the time IT staffs analyze all the server log files, PHP script, Java and .Net code, and database transactions that may have caused an outage, "users are irritated, the app still isn't working, and I may have advertisers or customers who are losing money because of it," Baum says.
In its first year with a product, Splunk signed 125 customers. Apple uses it to troubleshoot Web apps and maintain the e-mail system it uses to communicate with customers. FedEx bought Splunk's software to help manage Sarbanes-Oxley compliance. The U.S. Postal Service has used it to cut by nearly 60% the average time between crashes for tens of thousands of servers.
Splunk is poised to compete with IBM, Hewlett-Packard, and other large vendors of software for managing data centers. Baum says the IT knowledge he and his management team gleaned while working at places like Apple, Microsoft, and Yahoo helps them better understand the hodgepodge of systems large IT shops reckon with daily.
As for his time at IBM, it didn't last long. Baum arrived in Florida at the tail end of IBM's legendary PC project, working on the system's BIOS and other boot-up software. "It was a gas. Everybody knew they were doing something special," he says.
That was then, though everyone agrees that a well-tuned Web is something special, too. Baum's doing his part to make it work.
Return to Innovators & Influencers
You May Also Like
The Total Economic Impact™ Of Fortinet NGFW For Data Center And AI-Powered FortiGuard Security Services Solution Study
Protecting Your Hybrid and Hyperscale Data Centers
Integrations to automate your framework compliance: ISO 27001, SOC 2, and NIST CSF
10 Considerations to Building Hybrid Mesh Firewall
Hybrid Mesh Firewall: An Essential Solution for Today's Distributed Enterprise