Only 20% to 50% of in-house server capacity is actually used, even with virtualization gains, according to numbers from MIT-connected startup Jisto. The company says it has a solution, though, which will save enterprises money.
The problem that Jisto is looking to solve is that, although companies usually provision plenty of cloud and in-house server space, artificial static walls, which are created with ownership profiles and resource groups, create waste. Servers are underutilized.
It isn't just the in-house servers, either. Cloud capacity is also redundancy-prone.
Large customers are only using half of the capacities they pay for, according to an ElasticHosts survey that was quoted on Jisto's website.
But despite the capacity, workers often can't get access to servers when they need them—the space is allocated to someone else, who may not be using it. That's a waste of money and a productivity issue.
Jisto claims it has developed a way of elastically optimizing server space so it gets used better.
The startup was recently profiled in an article by Eric Brown, of MIT's Industrial Liaison Program on the MIT News website.
Jisto says it's fixed the problem by using Jisto-managed Docker containers, along with automated real-time deployment, monitoring, and analytics algorithms.
"As the resource utilization profile changes for each server or different parts of the network and storage, Jisto elastically scales its utilization in real-time to compensate," Brown writes.
It's the "real-time" that makes it interesting.
Many enterprises tend to overprovision, Jisto CEO and co-founder Aleksandr "Sasha" Biberman said in the MIT interview. That's why there's always plenty of resources.
Yet despite that, workers complain about a lack of access to computing resources. That's because of the "common practice of splitting access into different resource groups, which have different levels of access to various cluster nodes," Brown writes.
Jisto's system claims to break down these allocations and provide a more dynamic environment.
"You can still have priority during your server time, but if you don't use it, someone else can," Biberman explained to Brown.
That means users can sometimes get access to more servers than they were allocated, and productivity improves because people aren't waiting around for their turn.
"If there's a mission-critical application that generates a spike we can't predict, we have an elastic method to quickly back off and give it priority," Biberman says.
And that elastic nature of his product is what Biberman says makes it work better than anything else. Automatic real-time decision making sends the job to the available resource. That makes it efficient.
"If there's an important memory transfer happening that requires a lot of bandwidth, Jisto backs off, even if there's plenty of CPU power available," Biberman told Brown in his article.
As soon as something changes, Jisto decides whether to stop the workload, pause it, or reduce resources. "Do you transfer it to another server? Do you add redundancy to reduce the latency tail?" Biberman says.
With his system, "people don't have to make and implement those decisions," he adds.
This article is published as part of the IDG Contributor Network. Want to Join?