Dismiss
Innovations
Everpure Simplifies Enterprise AI with Evergreen//One for AI and Data Stream Beta
Accelerate the transition from pilot to production with benchmark-proven performance, automated data pipelines, and a flexible consumption model.
Dismiss
June 16-18, Las Vegas
Pure Accelerate 2026
Discover how to unlock the true value of your data.
Dismiss
Innovation
A platform built for AI
Unified, automated, and ready to turn data into intelligence.
8:26 Webinar
Consolidate Workloads and Save: Shared Infrastructure Done the Pure Way
Consolidate workloads with confidence and lower costs, enhance array utilization, and simplify management, all without dedicated arrays.
This webinar first aired on 19 June 2024
The first 5 minute(s) of our recorded Webinars are open; however, if you are enjoying them, we’ll ask for a little information to finish watching.
Click to View Transcript
Welcome. I asked Chet GP T 40, what is pure storage known for this is what it spit out. Number one, all flash storage. Number two, simplicity and ease of use. Number three, data reduction technologies, number four, performance and reliability. Number five,
innovative software. All of these are dead on and serve as the fundamental basis of what you will learn today. I'm going to tell you how to consolidate workloads and save. This is shared infrastructure done the pure way. My name is Juan Mojica and I'm the director of Security product management here at pure
storage. By leveraging secure application workspaces. You can save on the cost of dedicated equipment, you can save on the power and rack space that equipment would take and you can service multiple teams working on containers and A I on a single share platform. Now there's multiple innovations.
The pure platform is bringing to market in this secure multi-tenant capability delivers what you've come to expect from pure innovation. Also with its ability to restrict access. This capability addresses cyber resiliency to lease privilege. It can help minimize the attack service which limits the damage
from a compromised administrative account the public cloud has fundamentally reset expectations on how quickly services like storage can be spun up. That's the challenge for anyone providing or managing a private cloud really differentiating themselves from the public cloud.
So how do you provide a differentiated service? Some ways you can do that is by improving in one of these four areas availability. Can you provide better up time? That's better than the public cloud and the poor knives, they claim performance. Can you optimize bandwidth latency resources for your application security?
Can you isolate your application simply and easily lastly observable. Can you understand your applications and your environments better and understand what really is required for them to function properly today to deliver a high quality service? The simplest thing to do would be just to deploy dedicated hardware for application, the dedicated hardware,
the performance is allocated just for that application. So check that security can be applied specifically just for that application and with pure, we're bringing our 69 guarantees with evergreen one. So you have availability as well. But there's an associated cost that comes with dedicated hardware.
Another option is to share the hardware across applications. There's always apprehension when placing multiple applications on a single platform. Can you avoid the noisy neighbors from taking down systems? Can you delegate administration without giving one application team control over another? Can you clearly account for what each application is consuming?
Typically, these shared environments are considered to have a lower service quality. The question is how can we have that high service quality and maintain it but have the implementation costs lowered? Let me give you a scenario. Have you ever tried to deploy applications onto shared infrastructure and then try to allocate
costs to each of these applications? So you had three different DEV ops teams working on new A I initiatives. They're using containers and Kernes and they need persistent storage and went ahead and deployed port works. Now you can charge based on provision capacity but even the cloud providers are starting to
move away from that really shift to kind of a different way of uh making money with pure. You're getting data reduction across all your data on the array. So how do you allocate costs when you have data from one application being reduced with data from another application?
Can you tell how much space any of these workloads in isolation would consume if they were to actually have a dedicated system? Do you have the observable? You need to really run a service efficiently pure is bringing a market, secure application workspaces with the observ ability required to run a true cloud platform out of the box.
You can see what the consumption of a workload would be if it were deployed on its own dedicated equipment. The observable takes into account data reduction. Like if this were the only thing on the array, if it were on its own array it would have data reduction applied to it and that's actually accounted for in our space reporting in the
footprint of the application. This is the most equitable way to report consumption in your private cloud. You can also find workloads that in isolation would actually be more expensive to operate because they're constituted of a bunch of nonreducible data by themselves. Not only do you have the observable, but you can delegate provisioning and
cap consumption from both a performance and a storage perspective to prevent an overzealous K. Benetti's A I team from taking down others in a shared environment with our innovation. We're delivering a secure multi-tenant platform with secure application workspaces where individual teams have their
own dedicated resources and maintain their autonomy with supervisory guardrails and with the pure ethos transitioning to using this functionality is non disruptive, existing, non disruptive operations like upgrades remain non disruptive. Now, this is shared infrastructure done the pure way.
I would highly encourage you to join the webinar on July 10th to learn more. And there will be a test drive module for you to test this feature out once we're closer to release time and this lab would allow you to test out the secure application work flows to actually deploy multiple tenants on a single shared platform. So please attend the webinar and I wanna thank
you again for joining me today. Have a great rest of your day. Take care
Juan Mojica
Director, Product Management, Everpure
Consolidate workloads with confidence and lower costs, enhance array utilization, and simplify management, all without dedicated arrays. Learn how to enable FinOps at the workload level with a unique granular visibility to accurately calculate costs for capacity planning and determine the profitability of workloads. Explore how to easily manage your shared infrastructure or multi-service environment with four times fewer parameters to provision.
We Also Recommend...
Personalize for Me