Intel Links Up With Scale Computing To Sharpen HCI At The Edge
“With this product, I can start a conversation with a customer about hyperconverged infrastructure, that doesn’t end at ‘$100,000,’” says Scale partner Mark Essayian, president of KME Systems, an MSP in Lake Forest, Calif.
A server not much bigger than a paperback made by Intel with Scale Computing software inside can be used to put a hyperconverged infrastructure (HCI) environment at the edge with less hardware and fewer people needed to deploy and manage the set up.
With Indianapolis-based Scale’s HC3 operating system installed, Scale Computing CEO Jeff Ready said the NUC EEC product announced this week at Intel Vision is ready-made for forward environments in factories, retail, or restaurants. Those are areas with poor bandwidth, which is compounded by the mounds of data produced by connected devices. That combination can quickly overwhelm a network’s computing power.
“What they end up doing, is they’ll deploy something like (NUC EEC), physically, six feet from the robot that is generating the data, and just make a mini network,” Ready said of one use case for the server his company developed with Intel. “The robot is dumping all of its data and applications to the NUC EEC. So you’re processing that data locally and then only the output or the metadata or whatever you need from the output of that data, goes to the server room, or the datacenter, or the cloud. But you are capturing all of it there.”
Scale Computing partner Mark Essayian, president of KME Systems, an MSP based in Lake Forest, Calif., said his shop not only sells and deploys Scale products, but also uses three of its NUC servers in the office.
“MSPs need to look at this product, because they don’t need to pay for higher-level engineers to run it,” he said. “It takes 60 minutes to set up. That’s from opening the box, to up and going. Jeff (Ready) has put together a meal. I don’t want ingredients and a recipe. Scale gives you a menu of meals that your client wants.”
The Intel NUC EEC product is made for uncontrolled, non-I.T. environments and managed centrally through the Scale Computing user interface. There, admins can run applications anywhere they are needed, while also making less work for I.T. teams, Ready said.
“Some of this is the blue-collar reality of I.T.. Talking about the containerized application you want to run on-prem. These are the sexy things people like to talk about. But the other side of it, and I think our roots in the mid-market give us a very deep appreciation for this, a lot of I.T., we still have to make it work,” Ready said. “That grocery store that may very well have containerized applications and video surveillance and everything. They also have a point-of-sale system which is 37 years old. It’s not cloud native. They still haven’t ported it off Windows 2011. It’s hell and gone from this magical world. That’s just I.T. That is a reality. It’s not a clean slate. You have to be able to deal with those legacy applications as well as the new modern stuff.”
Ready said this is the sweet spot for partners who embrace the edge, which he called a “hardware renaissance” for solution providers.
“As customers need to deploy on-prem infrastructure, they need help to do that, and that’s why the I.T. channel existed to begin with right?” Ready said. “I can sign up with an MSP or I could sign up with AWS myself, but either way, I’m getting a similar experience. On the other hand, Amazon isn’t going to help me stand up my mini-network on my factory floor, but by God my local partner can.”
Genito Isabella, vice president of system’s integration at Cat-Tec, Inc., an Ontario, Canada-based solution provider who has been a Scale partner for two years, said partners should take a “hard look” at the new Scale line up.
“All the propeller heads like me, we are so used to seeing all these big boxes, so when they get in front of Scale, and they see these three little boxes, they’re like, ‘There’s no way this can support the network’ “ he said. “I was the same way. But when I look at the numbers on my deployments … most of these are at 35-percent utilization. I create more problems for them when I do back ups, more utilization, than when they are doing their job.”