Home>Technology>How NASDAQ harnesses hyper-converged infrastructure and virtualization for zero downtime (VB Live)
Technology

How NASDAQ harnesses hyper-converged infrastructure and virtualization for zero downtime (VB Live)


Hyper-converged infrastructure is great, but adding virtualization as an integral part of your stack means more savings, increased IT agility, advanced multi-cloud ability and more. To learn more about how virtualization can remove complexity and increase efficiency, and how to choose the right HCI tool, catch up now on this VB live event!

Access for free on demand right here. 


Hyper-convergence really is a different paradigm in how we deploy data center infrastructure, says Mike Wronski, principal marketing manager at Nutanix, different from the traditional three tiers, or a storage array, some compute, and networking in between. By bringing storage and compute together in more of an appliance, turnkey model, it’s a way to simplify using standard off-the-shelf hardware.

“But we can go further,” Wronski says. “We don’t see that as the endgame of simplifying data centers. We see it as the foundational first step.”

For Bhavani Yellapragada, VP DevOps at NASDAQ, agility is a huge factor since the company combines enterprise-level offerings and stock exchanges with innovation, combined with the growth rate that the company has experienced.

“I’m in application development and DevOps, so I look at the world of infrastructure automation through the lens of application deployment automation,” she explains. “When setting up environments, having consistent environments that scale is a big factor. The opportunity before NASDAQ is large, and we need to be able to respond very quickly to these market opportunities and the changes in the market. Automation through the vertical stack, all the way up to application deployment, is extremely important to us.”

The other factor is that the company has a huge number of distributed offices, and every office requires an IT center, and that means tapping virtual desktop infrastructure, she says — and you’re immediately thinking of the use case for hyper-converged infrastructure there.

How do the two technologies converge? Around 2006 or 2007 there was a huge hockey stick uptick, Wronski says, where the original use case of virtualization was really noted, and was leveraging the maximum amount of your hardware.

“The nice benefit from that, and it was an easy-to-understand use case — you could buy less hardware or get more workloads, more virtual servers, on a single piece of physical hardware,” he says. “There was almost a gold rush in the virtualization industry over that time.”

Everything in the beginning was just basic management: Creating a new machine, giving it storage, giving it memory. And since then, all of the tool sets and things that were required to manage the greater data center have continued to expand. The virtualization companies providing that software have added new and interesting tool sets above that, from basic operational management, to analytics, to automation, and suddenly we have this huge complex stack of software that was running on a very complex stack of hardware.

“HCI got us the first step past that: Let’s simplify the hardware side, make that management easier, almost getting rid of the storage component,” he says. “We solved one of the major complexity problems in a data center there. And how I’ll tie that back to cloud is, all that complexity is expensive.”

Complexity can be mapped to man-hours or vendor training or software licenses. A lot of complexity, a lot of cost — but the cloud is a different model entirely, with a consumption model that’s much more fractional or pay as you go. And HCI gets us a little bit closer to putting on-prem and cloud to a more parity model.

But don’t just assume the Hypervisor, he’s quick to add. It’s all of the other things around virtualization that IT has to use to drive and deliver their full service, end to end, and the management and automation of the tools around that.

“If we can get the Hypervisor and virtualization tier converged more, take the complexity out of that, I would argue that we’re now starting to make our on-prem data centers look a bit more like a cloud,” he explains. “We can start to do things like focusing on service offerings, as opposed to being very infrastructure, hardware focused. I want to offer a compute service, just like the cloud, to my local users. I want to offer storage as a service. I want to offer even getting into automation and these next-gen new technologies. Containers and Kubernetes. All of these can be very service-based. Even databases can be offered as a service if your infrastructure is flexible enough.”

The old way we built data centers was fairly rigid, Wronski says, and IT became very project-based, which is very different from the cloud model, where you ask for compute, and you have some parameters around what you think you need for your performance.

“HCI gives us a step there; converging virtualization gives us a step there,” he says. “But if it’s truly a single integrated platform, now the same management plane and the same stream of metrics and telemetry is from virtualization, from hardware, from storage, and then maybe some app intelligence as well. When we combine virtualization with HCI, now we’re getting better operational intelligence, so we can start solving the problems quicker and not spend a bunch of time wasting time with silos and finger-pointing.”

For NASDAQ, Yellapragada says, the biggest impact of combining HCI and virtualization is that the company’s infrastructure has been able to keep up with the proliferation of new applications.

“It’s very agile, it’s very secure, we have been able to scale at our pace,” she says. “It’s a perfect bridge between your traditional data center and the cloud. With the way we manage our acquisitions, the way NASDAQ is expanding across the globe, HCI has been a real blessing to set up the virtual desktops. The impact has been huge. We can stand up a data center in less than weeks actually, as compared to the very traditional months that it would take. We’ve been able to provide facilities to staff extremely quickly. And I cannot repeat myself enough about the consistency that it all brings. It’s reported zero down time.”

To learn about what needs to consider in choosing a virtualization solution, the challenges of moving operations, how traditional infrastructures can coexist with HCI-based infrastructures, a deep dive into the technical nitty-gritty of HCI, cloud, and more, don’t miss this VB Live event!


Access for free on demand here.


Attend and learn:

  • How virtualization platform options on the market compare
  • Why to consider changing virtualization solutions
  • How to choose the right virtualization solution
  • How to save thousands in virtualization licensing costs

Speakers:

  • Bhavani Yellapragada, AVP of DevOps, Nasdaq
  • Mike Wronski, Principal Marketing Manager, Nutanix
  • Stewart Rogers, Analyst-at-Large, VentureBeat
  • Dilan Yuksel, Host, VentureBeat

Sponsored by Nutanix



Source link

Review Overview

Summary

Leave a Reply

Your email address will not be published. Required fields are marked *