Nomad_Soul - Fotolia
Chris Drake, an analyst with Current Analysis Inc., in Sterling, Va., detailed a new initiative, dubbed the Zero Outage Industry Standard, or ZOIS, which aims -- not surprisingly -- to establish specifications that eliminate unplanned downtime within enterprise networks.
ZOIS is backed by 11 major companies, among them Cisco, Dell EMC, Juniper Networks and SAP. With downtime costs now topping $9,000 a minute, according to some estimates, companies are searching for ways to reduce their exposure to system failures, Drake wrote.
Zero downtime is one part of the initiative; the other is finding ways to boost productivity and tap into commercial opportunities fueled by next-gen equipment that supports a zero-outage standard.
ZOIS has a long way to go; it hasn't released any timetable, and its list of zero downtime members is yet to include heavyweights, like IBM and Oracle, or major internet service providers. Drake said the association is expected to provide more information about its intentions after a Dec. 22 board meeting.
Read what else Drake has to say about zero downtime.
Moving fast, when you should move slow
Another thing engineering managers need to worry about: "network incrementalism."
Gartner analyst Andrew Lerner dissected the dangers safe and short-term incremental changes can pose in an era of disruption in his most recent blog post. Indeed, taking this approach will hamstring enterprises as they attempt to transform their networks into the agile resource their businesses need.
Phasing out incrementalism in favor of more pragmatic, long-term strategic thinking takes time, but it must be done, Lerner wrote. One tool that can be used to spur action: the use of unrealistic goals. While these goals may be unattainable, Lerner said, they force unconventional thinking, compelling network teams to take a fresh look at the problems they face.
Lerner went on to suggest a few examples of so-called stretch goals, including one that would require any approach to be built on standard x86 platforms. The ultimate goal is to create a highly available network infrastructure, capable of meeting new business and technological requirements.
Get more information about Lerner's thoughts.
Unlocking the value of the hybrid cloud
Torsten Volk, an analyst with Enterprise Management Associates Inc., in Boulder, Colo., examined what it takes to unlock the full value in a hybrid cloud in a recent blog post.
In an ideal world, Volk said, customers would be able to match the needed infrastructure with the requirement and dependencies of each application. Instead, enterprises find themselves running into three challenges. The first is they design their hybrid clouds to be too complex, relying on overprovisioning instead of focusing on the specific networking components required.
Second, they make too many assumptions, relying on Amazon Web Services or Microsoft Azure without taking into account performance limitations. And, finally, enterprises are finding their mission-critical applications are now locked into an infrastructure that may or may not support their needs.
Learn what Volk has to say about what it takes to engineer a hybrid cloud.
Upgrading when downtime isn't an option
Diagnosing why data centers crash
Picking the right cloud stack technology