We use this solution primarily for production data for our virtual environment. We run a lot of SQL out of it.
We use the on-premises deployment model.
Pure Storage FlashArray//X is the world’s first enterprise-class, all-NVMe flash storage array. It represents a new class of storage – shared accelerated storage, which is a term coined by Gartner – that delivers major breakthroughs in performance, simplicity, and consolidation.
Pure FlashArray X NVMe is also known as Pure FlashArray//X NVMe, Pure FlashArray//X, FlashArray//X.
Download the NVMe All-Flash Storage Arrays Buyer's Guide including reviews and more. Updated: November 2021
Fremont Bank, Judson ISD, The Nielsen Company
We use this solution primarily for production data for our virtual environment. We run a lot of SQL out of it.
We use the on-premises deployment model.
Speed has definitely been a big improvement for us. We were running a bunch of iSCSI to EMC VNX and that had a gigabyte bottleneck. Now, since we're running through a true fiber channel to the Pure array, we're getting 32-gigabyte bandwidth. That means the speed and accessibility for our users and our customers have definitely improved.
The Hypervisor that ESX and vCenter use, as well as the Orchestrator for some automation, have helped to improve my organization. There's a lot that VMware does for us. Probably 95% of our infrastructure is built on VMware's platform, hosted on-premises. This is soon to be 100%. We have some physical stuff that we're converting and VMware has made it possible to pick up and drop those servers onto their platform. We'll reach 100% by the end of the year for sure.
We do have VMware analytics stood up. It's a simple Linux machine that runs a Kubernetes container that talks back to Pure1, which is their public website for support as well as for analytics, which we're using also. It's just a simple API.
We also use vRealize. It helps us pinpoint issues as they come. We haven't done a whole lot of the automation through vRealize, but we're probably going to work towards that so that if we have an allocation issue, it can automatically shift things for us. But the DRS and HA kind of do a lot of that for us inherently. There hasn't been a huge drive to do any of that quite yet.
The duplication algorithm allows us to get a lot more use out of less storage. We're running a five terabyte array right now and we're running probably about 30 terabytes on it. So the duplication rate is pretty phenomenal, without a cost to performance. It still runs pretty smoothly.
They can tout the functionality and cutting edge technology that they have, but that's where the price tag comes in. The cost is high, but I think as they grow their business and get more customers that it will probably go down a little bit.
I think it's a lot better than what we were running for production as far as the VNX, which has fail drives about every month. It had the EMC VNX protocol. And it's not the VNX product itself, it's just that we have a much older one so it has older disks and drives that will fail.
We absolutely see this working as our company grows. Even though it is fairly simplified in the way that they do their RAID and everything like that. It makes pulling drives and putting new ones in super simple. The costs could be improved though because it is quite expensive.
Their customer service is phenomenal. In their escalation paths, everything is laid out in black and white and it's very streamlined. Then they even go so far as to actually talk with other vendors on those integrations. If there's a potential issue with VMware and there's a potential issue with Cisco, Pure engineers will actually talk to Cisco engineers or VMware engineers and come to a resolution together instead of pointing fingers and saying, "Well, it's their problem, not ours." I've actually seen that in action so it's not just talk. You can actually do that.
It ended up being an incompatibility on the Cisco end for the UCS from one of the drivers that we had installed based on the firmware that we installed for that particular blade. But rather than Pure just saying it's Cisco's problem, call Cisco, they actually got on with a TAC engineer and talked us through it. With this method, they came to a resolution a lot quicker than it would have been to just open a ticket with Cisco and start the troubleshooting process all over again.
I had worked at a different company and we switched from EMC's XtremIO flash array to Pure and we were actually one of Pure's first customers when we did that. Their simplified support model and then their Evergreen program where they upgrade controllers as they come out was pretty phenomenal. I carried that over into the company I work for now. I kind of suggested that as a route to go in.
We used VNX, which we still have. I think one of the main driving points at the time was that we were running out of space on the VNX and the flash pool, and we could have added additional space, but the cost comparison between getting a Pure array and adding a new shelf to the VNX was about the same. I felt like we would get more bang for our buck going to Pure, which we have.
The initial setup was very straightforward. It's pretty much plug and play.
We actually used Pure professional services, which we didn't even really need, but we used anyway because it came with the implementation purchase. We did use a reseller to buy. We used RoundTower to buy the Pure array. We used professional services from Pure. It was probably set up within 20 minutes ad we could've done it ourselves, but it's always good to have trained hands on it.
A SQL job that would normally take six minutes running on VNX takes two minutes on Pure. That's been pretty beneficial because we're constantly running reports out of those SQL instances. We've seen enhancement and efficiencies from that.
EMC still has good products so they were still there. We were looking at adding additional space to the VNX. And then for general purpose, we did look at some stuff through Cohesity, but we needed more performance-oriented space. So that's why we went with Pure.
Try to get as many discounts as you can if you go with Pure.
If you don't need performance-oriented space or storage, Cohesity is very well priced and scalable. If you just need file server storage, you might not want to go with Pure. You might want to go with something a lot cheaper or more cost-effective like Cohesity.
The Systems Engineer that we worked with was a huge help, too.
I would rate this solution as a nine out of ten. It's very functional and very cutting edge. Technology just costs a lot.
We use this solution on private cloud for all our Oracle workloads and databases, and also for our enterprise virtualization infrastructure running BMR.
The main reason why we run our VMware on Pure was because it went hand in hand with the ease of use, configuration, compatibility, and the support. It was a recognized solution very early on and Pure Storage and VMware have a great partnership. It offers support for compatibility drivers and things like that.
We also use some of the integration tools from Pure Storage to help us manage some of the day-to-day tasks that we do with our storage and configuration and also to help keep things running optimally. One of the things we're using is the management pack from Pure Storage for vROps, so that we can continuously monitor the entire environment as well as our Pure Storage. With that we are more proactive about being able to make adjustments to meet the demands of our users. We use that to automate some of our storage augmentations when we need extra volume.
The moment we started using Pure FlashArray//X NVMe, we immediately saw a 30 to 40% improvement in performance. In addition to that, we saw a highly dense configuration where we were able to eliminate costly spindle-based storage and free up valuable real estate in our data center. That was immediately recognized.
VMware is really beneficial for our IT board. It benefits our IT organization by enabling us to consolidate as much of the physical hardware as possible within our data center. Essentially we can shrink the footprint to make more efficiency gains out of our existing data center.
What I really like about this program is that it is easy to use and easy to configure. I am also very happy with their support. It's the first storage company that we've actually engaged with and formed a partnership that does what they say they're going to do and they meet all of the requirements given.
Every time I think of something that needs to improve, they're one step ahead, which I love. The only area I wish to see improve, I believe is coming, is in the FlashBlade product. Blade implementation fell short on a few of the services.
Pure is an innovator. They're a disruptor in the industry and I love that. That means that they're motivated and dedicated to being at the forefront of technology. So the only thing I can say is that I wish that Pure would keep up the good work and it looks like they are.
We've been a customer for about four years and as far as the stability goes, we have not had any issues with the product. Again, it has met the described functions that we need. When our vendor told us what Pure Storage does, we weren't certain of that because many vendors make empty promises. But Pure Storage was the first company that delivered what they promised.
As far as stability goes, we have yet to have an outage. We have had failed components, but we've had top-notch support, and we've been able to resolve critical issues within hours. So if I had to rate the stability with Pure Storage products and our environment from one to 10, I'd give them a 9.7.
For our environment, we have seen the ability to scale up and out very easily with the product. We have also performed storage and controller upgrades live with zero downtime. That is another key component of our requirements that were met. I have been involved with every upgrade and there have been no issues. Scalability for us has been great. It works well and it does what it's supposed to do, as advertised.
What I like about Pure Storage technical support is that when you enter a request, you immediately get a response. Within less than 15 minutes you get a follow-up phone call, depending upon the severity of troubles you're encountering. That has been very good for us. And again, if I had to rate that on a scale of one to 10 I would say that's about a 9.5. That too is another thing in the industry that we haven't experienced before.
We were running on very old hardware that was all spindle-based sand solutions. Large-sized equipment with too much power consumption. We looked at other solutions like Kaminario, Nimble, Nimbus, 3PAR, and Tintri, but choosing Pure Storage was a no-brainer. We saw the solution, we were skeptical upfront, but when we did our research, it was very quickly recognized that Pure Storage was the right solution. We were able to save a ton of data center space with density ratios we've never seen before with hardware.
With Pure, we immediately saw value in the density ratios that we would be able to achieve. We saw value in the performance, in the scalability, and with the support. In the past, we've had trouble with support with all the other major vendors. So that was something that we took a risk on. Thankfully we did because it worked out.
The initial setup of any storage solution is complex. However, what Pure Storage has been able to do, is remove those complexities. When you receive the product, you will get detailed information on what to do. It really does eliminate the complexities. In other words, our experience has been very positive with the product out of the box. We did not need dedicated support engineers, we were able to figure these things out on our own, and we did it based on their documentation. If I had to rate that on a scale of one to 10 I would say that it's about a 9.5 as well.
One of the interesting things about ROI is that most companies assume ROI without quantifying it, which has one of the beauties of where I work, we were able to quantify an ROI. We looked at historical data from our previous spindle-based solutions and year after year we looked at those costs, the performance, everything down to the movement of water that it takes to cool this equipment. So with Pure Storage, we did a comparison in the same manner, and we saw a dramatic improvement in precious real estate space. If you had to put a dollar value on it, we'd saved a lot there. Don't expect ROI within the first year. Give it about two years and then you will start seeing a difference.
As far as the licensing costs, everything is included in the license. That's something that our other competitors don't seem to do very well with as they usually have additional costs.
On a scale of one to 10 Pure Storage has rightfully earned a 10 rating. We are very proud to be a customer and I hope the developers will continue to innovate and keep up the good work.
We are using this solution for our on-premises, private cloud.
The most valuable feature of this solution is reliability.
The UI for this solution needs to be improved.
This product is very stable and I have had no issues.
This solution is very scalable. It's good enough for us at the moment.
I have had no issues with technical support.
The initial setup of this solution is straightforward.
We used an integrator, WWT, to assist us with our implementation. We have had no issues with them.
Our licensing fees are $500,000+ USD.
VMware benefits our IT organization because of its ease of deployment and manageability.
We are using VMware on Pure, and we have implemented this because of the speed. VMware is faster, so it helps when you log in.
We use vCenter integration. It helps because you can see the storage added, and manage it from vCenter with a single console.
The cost for Pure may be high, but the reliability and scalability make it well worth the money.
I would rate this solution an eight out of ten.
We use this solution for our primary storage. Our deployment is in a private cloud environment.
Using this solution has benefitted us operationally by making us more efficient.
VMware benefits our IT organization through cost efficiency.
The most valuable feature of this solution is its ease of use.
I would like to see replication and DR features in the next release of this solution.
This solution is pretty stable.
We have not tested scalability yet, so it is to be determined.
Technical support for this solution is very good.
We did not switch from a previous solution. We brought this in as a way to improve operational efficiency, which is something that we're always looking for.
The initial setup of this solution was straightforward. It was an easy setup.
We used an integrator for our deployment, and our experience was ok.
We have not yet seen ROI with this solution.
We considered IBM and Dell EMC for our storage solution.
My advice for anybody who is researching this or a similar solution is to give it a shot.
I would rate this solution an eight out of ten.
We use the on-premise deployment model of this solution. Our primary use case is for machine learning analytics.
It has benefited my organization because it has reduced time to insights.
Its agility has been a benefit for our IT organization. We are running VMware on Pure. The main driver was FlashStack. The joint solution has helped my organization through its support.
The performance is the most valuable feature of this solution.
We have taken advantage of the VMware integrations developed by Pure with the validated design, FlashStack CVD. The integration has helped in the way that engineers feel competent that the solution is designed correctly.
In the next release, I would like to see real-time analytics for further insight into consumption models.
The scalability and telemetry analytics could be improved.
The stability is good.
It is somewhat scalable but it's not infinitely scalable. They could improve the petabyte-scale with greater capacities.
We haven't needed to contact technical support.
We knew that we needed to switch because our system wasn't performing so we knew we had to buy something. We liked Pure because it had a good pitch.
The initial setup was straightforward.
We did the deployment in-house.
We have seen a two year ROI.
With VMware, we pay $300,000 annually.
We also evaluated Hitachi, Nimble, and EMC. We went with Pure because they have the best pitch.
The advice I would give to someone considering this solution is, don't wait. Go for it.
I would rate it a seven out of ten. There is room for improvement.
We use this solution for on-premises, VMware workload storage.
I do not hire as much tech support for this solution as I would normally have to because of the way that it works, so it saves money on staffing.
VMware benefits our organization because we make use of the applications for everything.
The most valuable features of this solution are its ease of use and performance.
You can't use this solution with VNC, yet. We would like to see VNC integration or be able to use Pure Storage with VNC.
I would like to see the cloud version of the application improved. It is kind of new and it still has some limitations. Using it with VNC would be nice, and making it more incorporated with vSphere would be very nice.
The stability of this solution is extremely high. Very good.
Scalability of this solution is decent, and I would rate it a seven out of ten. We've recently become a VNC client, and you can't use Pure Storage with VNC yet. They are working towards it, but in the meantime, it prohibits us from using VNC along with some of the Pure features that they have.
I would rate Pure's technical support a ten out of ten.
We switched from our previous storage solutions because of speed. At the time, Pure was a lot faster. It was flash, and we were on spinning disks at the time.
The initial setup of this solution was simple and straightforward.
We used Logicalis, an integrator, for our deployment. Our experience with them was good.
We pay approximately $50,000 USD per year in licensing fees.
We looked at EMC, Pure, and IBM. The Pure interface was much better, which is why we chose it. The cost was also better in most cases.
We run VMware on Pure, mainly because of the speed and the ease of use of the Pure storage interface. The joint solution has helped because we can do the same thing with a lot less tech.
We use the vSphere plug-in for Pure. It helps because we don't have to learn the CLI for Pure. It allows my medium-level tech staff to do stuff that they normally would be a little nervous about doing on their own.
We are big fans of this product, and for anybody considering this solution, I strongly encourage them to use it.
I would rate this solution a ten out of ten.
We run all our Tier 1 and Tier 2 storage on it, our VMware infrastructure, all of that would be running on the //X70s. Our database workloads are on the new Xs.
The feedback I've had from the storage admins is that it's the simplicity. It's easier and quicker to allocate storage for us. We're able to get higher-density workloads on the same infrastructure, and we have a smaller physical footprint than we used to.
It helps simplify storage management a Database Admin perspective - there was a lot of thought that went into the size of the disks, how we allocate those, etc. Especially when doing maintenance or expanding disks. There was always performance issues when expanding disks on the old infrastructure arrays and allot of care had to be taken on offset sizes etc - Whereas now, we're finding that we can expand disks without having to remove and add disks again. That simplifies those admin tasks without any performance implications. From a Storage Admin side they love the ease of use and visibility the systems gives them.
Those are the three most valuable features that I've observed.
The visibility you have on what the frames are doing, even your phone is amazing, very detailed information about the environment in real-time.
One area that I haven't been sold on yet is the POD replication. We've seen that when we create a POD in synchronous mode, it increases the latency. We have another array that our latency isn’t affected by replication and when compared to the X array with a synchronous POD setup – it’s faster to respond to the system, but as soon as the POD mode is Async – the X is faster again. We not talking huge numbers – with the POD in synchronous mode with talking under 1ms for most IO operations.(Distance dependent to the other array) I do feel there is some gaps in my understanding of the POD setups in detail so perhaps its tuning gap.
The stability has been excellent, there was one minor issue when the M70s were first released – but they have been 100% stable since.
The only aspect that I can talk to is the benchmarking that I did from a Database perspective, as we purchased the X70 for our Database workloads. What we want to see is as we increase workload the latency should remain stable. We are able to push the X70 way past our current needs from a throughput and IOPS perspective – without any degradation on latency. As we ramped up more machines into the tests, the only limitations we ran into were switch ports and FCoE saturation – the X had headroom.
The technical support is very good. The teams are awesome. The guys in South Africa, and they're very, very good.
If you're looking at it from a database perspective, you obviously want to make sure that you can scale your workloads without a degradation in performance. What we've seen on the X’s, and the Pure’s in general, is that we can scale beyond what we need without any degradation in performance. The real-time monitoring I’ve seen via your phone is a differentiator alone – to get a heat map in the past was a big process and took place after the event.
In terms of the Predictive Performance Analytics, I haven't seen it myself. I know that the guys have just recently started looking at a lot of analytics.
Although I don't admin the solution, I would give it a ten out of ten - from the vendor technical expertise and helpfulness, it's simple, reliable, quick and predictable at the same time I’m expecting lower costs for us once we fully migrated.
I didn't do the setup, but everything I heard was that it was straightforward.
We used the main guys in South Africa, Data Sciences. They do everything, top to bottom if you want or with you teams. Our experience with them has been awesome.
I do know there has been a reduction in the total cost of ownership, although I can't say how much as we in a transition from our old arrays, but I do know the cost per gig is lower, the dedupe on the array helps drive this cost down – as well as the physical size is much smaller and uses less power.
The vendors at the time would have been Dell EMC and Hitachi.
It depends through which lens you're looking at it. If you're looking at it from a database perspective, you obviously want to make sure that you can scale your workloads without a degradation in performance, your latency times. What we've seen on the Xs, and the Pures in general, is that we can scale way beyond what we need to without any degradation in performance. We don't need to sacrifice any of the performance as we scale up or scale to the side. There aren't many vendors that we looked at that can scale to the size of the operations that we needed, from an I/O-testing perspective. In real, day-to-day, we don't run that hard. But if we need it, we can. It's there.
In terms of the Predictive Performance Analytics, I haven't seen it myself. I know that the guys have just recently started looking at a lot of analytics, but I haven't seen it myself.
Although I don't admin the solution, I would give it a ten out of ten. It's been awesome to work with. It's simple, it's very reliable, it's very quick. And we get excellent dedupe ratios on the machine without a sacrifice in performance.
At Secure-24, we have been a long time customer of Pure Storage. We started with Pure Storage approximately four years ago. We use it as the primary mechanism for block storage in our environment.
We specialize at comprehensive managed services of business-critical applications. We run a hosting environment and the full gamut of applications, infrastructure, security, compliance, and governance. Using that model, Pure Storage is a key part of being able to deliver the performance, encrypted storage, and reliability for mission-critical applications.
The biggest impact changing to Pure is we saw an influx of tickets from customers because they thought their BI applications were broken because they were running too quickly.
Operationally, we find that we are not having storage problems anymore on performance. Historically, there might be various storage performance issues when you have DBAs involved utilizing network and storage resources, since it's very operationally intensive to gather multiple teams together and do comprehensive troubleshooting of problems. We found these issues have simply gone away since we have migrated to Pure Storage. It's been fairly significant change in how we manage and deploy applications.
It's helped us because we've changed fundamentally what we talk about. We don't talk about storage and different tiers of storage anymore nor do we talk about servers. We talk now about applications and how applications impact the business and end users.
You don't have to worry about the different tiers of storage. They are always fast and reliable with consistent performance.
Based on the various types of workloads and environments we run, we have looked at Pure Storage as being one of the largest infrastructure-based changes we have seen in our environment in the last 10 years of infrastructure hosting.
What has changed is you now have databases and applications where you can make an infrastructure change and it directly impacts the end user experience. You generally don't see that with other infrastructure changes. If I change from storage A to storage B, maybe there is a small or minor performance increase.
With Pure Storage, there was a dramatic performance increase which we saw across various different applications. Going from a legacy vendor to Pure Storage, we saw reductions in MRP reports previously running at six hours going to 30 minutes.
With Pure Storage, we would like to continue seeing price reductions with flash storage. I don't think we're any different than anybody else when we continue to look to the industry for price reductions of both NVMe and traditional SSD storage. We would like to see these prices continue to decline and erode, even displacing large spinning disks.
Ideally, in a perfect world, you would have all-flash arrays being able to displace even your traditional "cheap and deep" type storage frames. We are more excited from the industry perspective when this type of transition can happen from a cost perspective.
Also, I want to see Pure Storage not only be for fast storage, but I want to see it be for the entire data center.
With mission-critical, business-critical environments, stability is of the utmost importance. Pure Storage has impressed us in this area. We have gone through multiple software upgrades, as well as completely non-disruptive hardware upgrades. Upgrading from an FA450, which is one of the arrays from more than four years ago, to an M70, then to the X70s. We have gone through that, the full controller and full storage upgrades, non-disruptively to any of our customers. This has been a big change from what we've seen historically in the storage industry, where you had to do an upgrade or when you had to do a large new purchase, and there would be a significant amount of time, planning, and organization which would go along with it. There would be intangible costs that would generally come along with legacy providers.
The way that we have looked at scalability is from a linear scalability perspective with multiple storage frames. We like having the capability of scaling wide with multiple storage frames as opposed to trying to scale too large with any one individual frame. However, we also have an X90R2 with two petabytes of NVMe in it which fits in about six rack units of space. This has been transformational, as well. From there, we scale out linearly with multiple of X90R2s, as opposed to trying to somehow cluster them or make them larger.
Over the last four years, we have displaced the majority of other storage solutions in our data centers with Pure Storage.
The engineers and architects on our team were the ones directly doing the setup and implementation.
We have absolutely seen a reduction in total cost of ownership (TCO), specifically around the operational overhead of running a storage array. Now, we have our Pure Storage arrays managed by VMware resources. Therefore, we don't have a specific role for storage management where one used to exist. So, the VMware team is able to manage, deploy, and configure the arrays. The simplicity of drives allows for this consolidation.
There are a lot of intangible costs, such as gathering teams together in war rooms to troubleshoot performance issues, simplicity of managing the array, growing the array, and going through upgrades. When you look at all the intangible operational expenses, in many cases, these will offset the capital investment of going to an all-flash array.
To say, "Pure Storage has simplified storage," is a bit of an understatement. The array has gone from having a PhD in working with it to effectively having a high school diploma. Anyone who understands anything about IT can run a Pure Storage array. It's incredibly easy to use and greatly simplified our ability to both deploy and manage our storage subsystems.
With the traditional Pure Storage array, you had very consistent low latency, but you still were in one to three milliseconds. Now, with the all NVMe arrays, it's a whole new paradigm of fast. You're actually measuring everything in less than a millisecond. So, with the I/O responses, your high bar is one millisecond. This is something you haven't seen in most traditional storage frames or even all-flash storage frames.
We generally always advise people to make the choice to go with Pure Storage because they won't regret it. We can evidence that a lot through our experiences of running massive databases and systems on Pure Storage today and prior experiences that we've had with it.