Cloud services, virtualization and personal computing devices are among the trends in government information technology that complicate networks and could be contributing to frequent disruptions, according to a new report.
Agencies with more complex networks are three times more likely to experience disruptions than those with simplified networks, according to a survey of federal network managers by MeriTalk.
Respondents -- 94 percent of whom say network downtime in the last year has hindered their agency’s ability to carry out its mission -- estimate they could save 18 percent of their IT budgets by reducing network complexity by half. At that rate, governmentwide IT savings would be $14.8 billion a year, MeriTalk said.
More than half (54 percent) of federal IT managers said their network complexity increased in the past year, and 68 percent believe their networks will become more complex during the next three years, the survey found.
Factors that contribute to greater network complexity include an increasing number of network users, the move to cloud computing and server virtualization and increased use of mobile devices, including those provided by employees under bring your own device, or BYOD, policies.
Respondents said moving to open, nonproprietary standards is the best way to simplify networks, MeriTalk said. Simplified networks are more reliable, faster, safer, easier to fix and provide more IT agility to support agency missions, respondents said, according to MeriTalk.
“The network is the road on which all government information travels,” MeriTalk founder Stephen O’Keeffe said. “Taking the road less traveled by – one with reduced complexity, interoperability, and diversification – will make all the difference, and is critical to agency performance and efficiency.”
“The Federal Simplicity Report: Navigating Network Complexity” was underwritten by Brocade, a company that provides network services to government and others. Researchers surveyed 200 federal network managers in May 2014, with a margin of error of +/-6.86 percent at 95 percent confidence.