Pure Storage FlashArray Reviews

We primarily use the solution for storage. 

The administration is very easy and quite minimal.

The performance is very good.

The installation is pretty straightforward. 

Technical support is good.

Beyond a certain amount of petabytes, you have to have a separate system. Basically, it's not infinitely scalable. 

We've been using the solution in two places. We have used it in Washington for six months and London for six months as well. 

We have found the solution to be stable. We've not had any performance issues and not have any stability issues with it. The performance is good. there are no bugs or glitches. It doesn't crash or freeze. It's reliable. 

The solution is scalable to an extent. Its deduplication ratio is quite good. You don't actually always need the full amount of physical storage, as opposed to the deduplicating amount of space.

Its scalability beyond a certain amount of petabytes makes it so that it is a separate system if you scale to that much storage. If you're a normal enterprise where you don't scale that much, it's fine. The X90 goes beyond a certain amount, and you have to have a new system that works as a separate system.

Across two locations, we have about 800 people. 

From the feedback I've gotten, for the most part, the technical support has been quite good. They know their product quite well and they've been quite responsive. We are satisfied with the support.

We previously used EMC VNX.

The solution is pretty straightforward. It's not too complex or difficult. 

I haven't personally deployed it, however, from the team's feedback, they had very little issues with regards to the deployment and bringing it up and then setting up. The deployment was within days or a week at a maximum. That was for the migration time. For migrating from the old storage to new, it might take a while, however, in terms of getting it up and running, it was days. It didn't take that long at all.

It's easy enough that the normal system administrators can take care of the storage. There's no one dedicated to managing the storage and it's not that complicated that it needs someone dedicated just to manage it all.

You do have a license that is required and you also have to pay for support.

The license we have is not a monthly subscription model.

For London and Washington, we had EMC VNX. At the time we were transitioning from the EMC to a mid-tier flash storage, and they didn't have anything in the market. We did look at others. We looked at the Nimble from HPE. When we were looking to purchase, EMC just came out with EMC FlashSystem. We didn't get to the IBM FlashSystem. 

We're a customer and an end-user.

We are using the latest version of the solution at this time. 

I'd rate the solution at a nine out of ten.

I would recommend the solution to other users and companies. 

We use the private deployment model of this solution. In terms of our cloud provider, we use Azure, we are signing on with AWS, and we'll be using vCloud in the next quarter.

It replaced an earlier tier. It replaced 3PAR Storage and gave us faster performance than the single databases.

VMware has benefited our IT organization because we're 100% VMware, everything is running on it.

We are running VMware on Pure. Our main driver was the performance for SQL servers. The joint solution has helped my organization in the way that the databases run faster. 

My organization is taking advantage of the VM integration developed by Pure. We've deployed it. I think it gives the storage administrator some additional insights on metrics. I don't think we're using it to actually manage the data stores. He's getting more insights on metrics. Pure has a VAAI plugin that allows you to manage the data stores. We're not doing that, but I think it gives them heightened analytics in addition to SD-Pure1, a web interface. The integrations have helped in the way that they're another dashboard to have. Somebody could think that the databases are running slow and our database administrator can look at that tool and say, "No, it's unique to your SQL databases, it's not the other VMs on the data stores."

The ease of management is one of the most valuable features of this solution. I would have also said that it's pretty fast but now our SQL servers are starting to beat it up pretty bad.

A year ago they promised that they would be able to read through the database encryption with more metric and they have not delivered on that patch, which is significant because it gives us back so much more storage room. We want to be able to read through the encryption.

It's very stable. 

It's scalable. You can hit a point where you fill up enough drives in the shelves. We're at that point now where we've got to expand. We've got to add another shelf.

Their technical support is good.

We have switched to EMC. They gave us more array for less money. 

The initial setup was very simple.

We used an integrator for the deployment.

Our ROI is that we're still running. It's been two years later, and we're still up and running with no downtime.

We looked at HP, NetApp, Pure, and EMC. EMC gave us better-performing storage for a better price. 

I would rate this solution a 7.5 out of ten. To get to a perfect ten they should be more competitive in their pricing. It's expensive. It's premiere storage but there are other premiere providers out there as well that are beating them on price, at least in our case.

The encryption is another area that needs improvement. It was huge. Right now we're at 82% on the Pure array. If they come up with that and pass to read through the more metric encryption, we would probably get 30 or 40% available disk space back, so it's huge.

We primarily use this solution for our SQL server in an on-premises deployment.

Having a dedicated array for our SQL server is very nice.

We are running VMware on Pure, and the main driver for that is because it is all-flash. Also, we wanted a dedicated solution for our SQL environment. Running on Pure has given us the ability to scale out our SQL environments. We tripled our environment in the past three years since implementing this solution, and we have not had any issues with the storage keeping up with the workloads.

We are making use of some of the VMware integrations that have been developed by Pure, but we are really waiting for the copy data management part.

We are really enjoying the speed of this solution. The amount of throughput that we're getting is really nice.

In the next release of this solution, we would like to see automated copy data management for SQL Server.

We have had zero issues with stability once it is in. However, we have had issues with migrations to different cabinets or different arrays. We had one instance with an eight-hour outage in our primary data center because the upgrade to the controller failed, and the controller redundancy didn't work. It was an odd issue that we now have under control.

This solution scales well. The issue we had with stability is now under control, so we are able to scale out fine. We can just drop in new disks when we need them.

When we've had issues, technical support has been really good about resolving them quickly. I was on the call with them when we had the issue with the controller, and they were very, very helpful.

Our older solution was not very good. Pure increased our speed a lot. We needed to increase our storage because we were filling up the array. Our SQL footprint has greatly increased over the past three years.

This solution was chosen because we happened to be doing a POC when our previous solution failed horribly, and we moved our production to Pure. It was able to pick it up, which was the selling point.

The initial setup of this solution was pretty straightforward. It was a vanilla, out-of-the-box setup with nothing out of the ordinary. 

We used an integrator to assist us with the implementation and deployment of this solution. We were hands-off, but it seems that all went well because everybody is happy with it.

We have seen a good return on investment, mainly because we took our SQL Server workload out of the general population and we're able to get it separated, which is a huge advantage to us. The biggest boost is getting separation of duty.

I have used InfiniBand in the past. We are now looking at building a new data center, and the vendors on our shortlist are Pure and InfiniBand.

We are now starting to look at some of the copy data management tools that come with the new array.

This is now my go-to product, and I was an InfiniBand guy before. I like how there are database integrators on the Pure team that are actually there to help you tune your database workloads with their solution. I don't see that in a lot of other vendors.

This is a good product and the overall day-to-day workflow within it is great, but some of the issues that we've had with migrations bump it down slightly. The product is good, but it could be better.

I would rate this solution a seven out of ten.

We primarily use the solution for desktop virtualization.

I have IOPS and IOPS input/output. The reason that we have virtualization required for the media is because of high IOPS and we're able to maintain it with PR. The encryption is pretty high. We like the encryption right on the storage.

I was able to put up more VMs using Pure. I'm running almost 3,400 VMs and VDIs on Pure Storage. This improves our organization because we can just set it up and we forget about it. Everything works. We do not need to worry about storage or bandwidth issues. Its ease of use is also helpful. The setup is very easy with Pure.

I'd like to see a move towards individual VMs for what the performance of each VM is in a VD infrastructure. I can see the overall volume, but I would love to see things in a more granular level on the VM side. I'd like to say "Hey, this particular VDI, what is the performance on that? How much IO is it using, what are the issues, what is CPU?" etc. I'd like to see that layout in the portal. That would be great for us.

The stability of the solution is very good. After five years, I've had very few problems. In terms of problems, for example, sometimes I've seen some spikes in iOS. It came from our end, not from Pure.

The solution is easy to scale. I'm running two environments right now, so I need to scale. I've got an A-side and a B-side.

The initial setup was very straightforward. I did the GUI configuration after Pure finished their end, so it was very easy for me to set up. They just did the back end. I did the physical setup. They came back and did the configuration on the heads and I did the GUI set up with the network configuration, so everything else we set up ourselves. The setting up volume was very easy.

Pure assisted us with the implementation. It was a beautiful experience because we had an older model on which the head had to be upgraded. They did it seamlessly. I had no drops in my VMs.

There has definitely been an ROI. In four years I've never seen another storage vendor that offers what's called an Evergreen solution. I should have my refresh next year, so I'm getting a brand new a controller with a minimal cost. By then we're going back and replacing the whole thing.

I did a POC with three different vendors. Pure won out due to its resiliency, adaptability and the IOS and the feature sets. I was able to pull up all three discs at the same time and it never failed.

We are using the private cloud deployment model.

We are running VM on Pure. The main driver around VM on Pure is the number of IOPS I was able to get out of the two controllers. That was the main reason I chose Pure.

I'm not using any plugin with the vCenter or anything else like that.

The advice I would give to others considering implementation is to do your investigation, do a POC, and try it out. Find out which fits your needs. Also, isolate your workload. Don't mix your workloads if you want to do a successful VDI deployment.

I would give the solution nine out of ten.

As a customer, we use them as our Tier 1 storage arrays. It has been amazing. It's extremely fast, reliable, and resilient.

We have done a lot of different things with Pure Storage. We have included some real-time analytics that we developed for our eCommerce website and run those on FlashBlade. We used FlashBlade as it was the only storage platform fast enough to keep up with that data flow.

We are able monitor I/O, latency, read/write, capacity used, and all the different metrics that the Pure gives us the ability to monitor.

It definitely affected the ability to capacity plan, but in a good way. We have all the visibility into the capacity, forecasting, and all the metrics that the solution provides us with.

It takes drastically less time to manage and administer the solution. We would have about three or four people who were dedicated just to work on storage with only one guy who could actually do the Hitachi replication, because it used old archaic technology called HORCM files. In the Pure Storage realm, this is not true. All our junior partners can administer the storage arrays. It is simple and easy to use. We don't have to dedicate a whole team of full time people to work on it.

When I was a customer, the most valuable feature is the ease of use. 

It is the whole package: The ease of use, cost, and the ability for it to perform at a level that traditional storage arrays just can't compete with.

It simplifies storage. In the old days, you had to go and decide what ports were going to go to what workloads, which was a lot of work. You had to set up replication. Now, everything is just a few clicks away. It is set up exactly like you would want it to be. That is what it does. It simplifies and optimizes the solution.

I would like to see some improvements on the FlashBlade side around the CIFS space support. I am not super familiar with all the different NAS protocols that they run on their box, but there could be some improvements made on SMB CIFS side.

Some of the FlashBlade protocols could use a little love. There are obviously some new enhancements. There is no dedupe on the FlashBlade. It is compression only. There is no replication. So, Pure is going to try to partner that product with ObjectEngine to bring in some of those features, and I'm not sure how all of that will work out. I'm not familiar with ObjectEngine yet, but we'll see how it goes.

The stability is great. We have had no issues. We have never had an issue or outage that has been related to Pure Storage.

We decided how big of a failure domain that we wanted to entertain. We decided to split three into what could have been one single controller interface system. However, at some point, if we lose 500 terabytes, what does that do to our company? Now, we have things like active clusters which mitigate a lot of these issues, but people still need to be wary about how they design their failure domains.

The support is great. The support has been amazing.

We thought we we going to go with the new version of Hitachi, and everything was going to fine: Lift, shift and replace with the new one. What we started doing was exploring the marketplace, then figuring out, "Is this the best option for us? Could it be simpler?" Because the Hitachi was a tank, but it was not simple to use. It performed very well, but it did not perform like an all-flash array does.

The analytics are great. Previously, we had Hitachi solutions, and it was very hard to understand what was happening with the array. One of the great things about the Pure Storage solution is you can instantly know just by logging in or checking Pure1. You can do it on your phone. Hitachi doesn't have anything like that. It's amazing that you can get this type of visibility from your storage array. All the analytics feed up into Pure1, and you can just see them whenever you want.

It used to be that people would buy Pure Storage arrays and they would use it for a single instance application, like an Oracle database. We never did that. We used the product to replace our entire giant Hitachi G1000 storage arrays. Everything that we had went to the Pure Storage arrays. We had three giant M70s that are now X90s which house everything the company was running when I was a customer.

The initial setup was very simple. They came in before lunch, and we had it up after lunch. Then, we were already starting to move workloads to it after that.

We have upgraded firmware controllers and physical controllers. It works exactly like they say it does, which is the best part. You don't even notice. Business runs as usual. You can replace a controller, it fails over to the other controller, and everything runs smooth as butter.

We used Sirius Computer Solutions for the deployment. They have been our partner and VAR for a long time. They know our environment very well and were with us every step of the way.

From a footprint perspective, we used to have big giant racks of storage on both sides of the data center. We would have to plan and have a hole where the future one would go. Now, we don't have to do that at all. They are just sitting in the rack right next to it.

We have a seen a reduction in TCO. It is definitely a cost-effective solution for us. We have seen ROI.

We have an Evergreen Storage subscription. We like it a lot. We recently upgraded from the M-series to the X-series FlashArrays. We used the Evergreen Storage solution and expanded our footprint.

We evaluated Hitachi, who was our current vendor. We evaluated Dell EMC for the VMAX and XtremIO. Then, we evaluated Pure Storage. 

We are also a NetApp customer, so we evaluated them. However, we don't run any block storage on NetApp, only files.

Do a fair evaluation. Be objective, look at the different technologies, and use the technologies. See what they look like and what you will to have to deal with when you're using the products. It's easy to make a decision based on bullet points, but it's hard to make a decision on actual use of the actual technology.

We are a Chef shop, so we integrate it into Chef and VMware, vRA, and vRO. We also use all of the plug-ins. The integration is easy, simple, and seamless.

For most of the workloads, the solution’s inline deduplication and compression has performed fine. We had a few workloads that were already precompressed, so when you put those workloads on top of a storage system that does compression and dedupe, they don't compress again. So, they tend to eat up a little storage. Therefore, we specifically targeted some third-party applications, like IDERA SQL Safe, and tried to remove them from the environment. This way Pure Storage could then compress and dedupe those SQL backup files.

We are from Texas. Power is like ten cents a kilowatt. Texans apparently don't care that power is cheap. From a power requirement, it definitely has used less power, but we didn't use that as a metric to look at.

Biggest lesson learned: Why didn't I switch sooner?

The primary use case is SAP. It work very well.

SAP is very important to our business. We are running all ERP solutions. Our configurations are run on-premise. 

It has easy implementation.

The most valuable feature is its simplicity. It simplifies the administration and backup.

The predictive performance analytics works well.

I would like to have support available in Spanish.

The stability is perfect. The reliability is 100% and the latency is always lower than 1 millisecond.

The speed is very high on SAP from running it on Pure Storage, and the power on SAP HANA is much faster.

The scalability is good.

I would rate the technical support as a ten out of ten.

The initial setup was straightforward.

We deployed directly with Pure Storage.

The ROI is very positive for the reductions in HANA.

We have seen a 5:1 reduction in Total Cost of Ownership (TCO). 

We are finding the TCO of flash to be lower than SSD implementations by 2:1.

It's a simple, robust solution, which is very stable.

Pure Storage is very good and quick for backing up SAP HANA.

The first year, we started out with one or five terabytes and it took what was 20 terabytes of storage down to less than one terabyte. Then we added another one and I think the total storage is five terabytes now.

The UI is pretty good.

It's absolutely a stable product.

It's scalable. We started with a very small storage array and now we're in a much larger one. I think we're up to 40 terabytes.

We only use tech support infrequently. We don't need to call them. It's easy to use, straightforward. Once it's set up, it does what we need it to do.

Price is about the only thing that's wrong with it. A little bit better pricing would be great. The client environment for a non-profit 501C3 organization makes it much harder for us to come up with the dollars and to cover the increased cost of hardware support, but we do like the way the product runs. It's perfect for us.

I would definitely recommend this product to a colleague because of what it can do. I've already done that. I've already referenced several other nonprofits, human service organizations, and long-term care facilities. We've spoken highly of Pure. For an organization, it can take storage from 40 terabytes down to five terabytes. It's excellent.

Our user base consists of 3,000 people but it takes just one person to manage it - ease of use.

The primary use case is for the data and storage that we utilize in our managed services. 

We also use it in the company. We localized it.

Deduplication works faster for our customers using this product.

It simplifies building out the storage.

The most valuable feature is its data reduction. 

It is very easy to use.

It needs to improve its price.

The product is stable. It works really well.

I have not used the technical support.

In most cases, we do the implementation because we are the integrator.

We are finding the TCO of flash to be lower than SSD implementations.

The price is too high.

Because the price is a bit higher than other products, the data reduction equalizes the price with amount of the data reduction.

Go for it. The product is great.