SolidFire Reviews

We have approximately 8,000 VMs that we had been running on our traditional storage system and it simply was not able to keep up with the workload, so we've migrated all that to the SolidFire product. Provisioning times have gone down and a lot of the random errors from different things that we've seen across time kind of all went away. It's made everything much more efficient. It has saved us time.

We do a lot of tear-downs and rebuilds in non-production environments, so those processes have been reduced to minutes. It's been tremendously beneficial for our development.

The most valuable features for us are their QoS, the scalability and the serviceability of the environment. Our ability to add nodes or take nodes out for service and the QoS policies we're able to wrap around volumes are all very helpful.

The upcoming release is supposed to have much richer VMware virtual volume (VVOL) support, which is something we're very interested in. For our particular environment, we also use the VMware Integrated OpenStack, and so our VVOL adoption is waiting on VMware because they have to update their VIO product, but that's definitely a direction we want to move.

It's been rock-solid. We have not had a single incident. We've not had any latency issues.

It is very easy to scale. We started with our non-production cluster. I think we started off with six nodes. It's now a 14-node cluster. That's a seamless process. It just worked. No down time, no service disruption, nothing.

I have dealt with technical support many times. They’ve been very good. What they tout is they only have level-three engineers; there aren’t the normal layers of trying to get to somebody who can actually answer your question, because the first engineer you get ahold of usually knows the answer. If they can't, they basically have direct access to the engineers and developers. It's amazing; it works very well.

We had some NetApp 8040s and 6220s, which we still use for certain workloads because the SolidFires only do block; they don't service our NFS workload. The NetApps we had were flash pools, basically spinning disks fronted by SSD. Even with that configuration, they weren't really able to keep up with our workloads, so we needed something that had a lot higher throughput, so we started looking at all-flash technologies.

At the time, we didn't feel the NetApp offering was as mature as it needed to be, though we didn't technically evaluate that. We looked at ExtremeIO, we looked at Kaminario and finally the SolidFires. The ExtremeIO was really expensive. The Kaminario seemed slightly better but we liked the scalability story around the SolidFires. We then talked to some other customers who had them and confirmed that they really did live up to what the marketing hype said, and that sold us. For our highly dynamic VM workload, it's what the platform was built for, and it was a really good fit for us.

The initial setup was very smooth and straightforward. Basically, you put an initial configuration on each of the nodes and then they form a cluster, and then as you add additional nodes, you make it a member of the cluster. Originally, we had done that using their GUI. The last couple of clusters I built, I used their APIs to do it; very quick and painless process.

Look at SolidFire. It sounds cliché but it's true. For us, it worked really well.

Getting predictability in our analytics for space trending, performance analytics. We use to correlate data with other tools that we have. If we get complaints about any kind of performance metric issues, whether it's storage related or something on the virtual side, we use it to pinpoint what the actual issue is. It has proved really useful for that.

Part of the reason we went this route was we did that storage design workshop with NetApp. So we went for QoS-driven design for our new array. It really helped us not only in delivering the service levels to our users, but also automating that. So it makes it a lot easier for provisioning. It also makes it a lot easier to guarantee performance for our end users.

Nothing I can think of that they don't know about right now. They're looking at making some of the custom widgets and reports a lot easier to deal with. They're heading down that direction already, so I don't think that's a big deal.

For example, the ease of use with the reporting. Right now it's not impossible, but you have to know Sequel. It's a little time consuming to get those customized reports in there.

About seven months now.

It's great. At first there's a little bit of a learning curve, but once we got past that everything is rock solid.

So far it's been great. We've have not had any issues. We've added some more data into it, it hasn't choked on it.

We've used them twice. They were great.

It was really easy. We had Professional Services delivery with it. We worked with the NetApp CI team to implement. That's about as straightforward as you can get.

It is fast. By default, its APIs expose pretty much all of its configuration items. On the ONTAP systems, we use WFA to expose the APIs, where with SolidFire, everything is pretty much out-of-the-box, so the customers like it. The main uses are virtual machine environment. This is internal, on a private cloud. In India on most of their workstations are on virtual machines, and those all are hosted on SolidFire.

SolidFire is one of the products that does have great APIs right out-of-the-box. It works great. The tools and the other stuff seem to work a little better right out-of-the-box than the ONTAP stuff does, C-Mode.

It's doing SAN, so that would be the major difference. We use NFS file storage much more than we use block storage. SolidFire is our only block storage offering right now. Honestly, we're kind of phasing block storage out, but it's filling that gap for applications that claim they need block storage and can't use file-based. That's kind of its role.

It is just filling the gap of the block client, because maybe 10% of our clients have to use block storage and have a good technical reason. The other 90% we've gotten on a NAS.

They could do a file-based NAS: SolidFire NAS-based. It's probably not its niche, but that is our direction, not to use block, and it's block. Solid state block is what it is.

If it was the same price as C-Mode and did file-based storage, because this is what our company is heading towards.

It is pretty low-maintenance for upgrades and support issues. We haven't messed with it much. We have set it up, and we have the capacity so we haven't added a lot to it either. I have not had any issues with it.

Setting up new clusters is pretty straightforward. ONTAP is great, and it is really easy to use and setup.

It is very stable. I forget we have it sometimes, because once we have it configured, it just up and runs. Plug and play, the GUI works and the APIs, customers can use them. Everything is kind of there, therefore, it is very low-maintenance.

We have not scaled it a lot, because there are some niche environments running including the virtual workstations. So, I don't know how well it scales.

Customer Service:

It is fine.

Technical Support:

That is a problem we have. When we call tech support, we have to open a tunnel to the SolidFire device, then tech support can get in and look at it. The tunnel keeps closing on them. So we'll open the tunnel, 10 minutes later I'll get a phone call. "Hey, the tunnel closed. Can you open it?"

The only way to keep the tunnel, and I think it might be an issue with our proxy on our end, but something is closing the support tunnel. I don't know if it is a NetApp issue. We haven't been able to fix it on our end, so I don't know if they can give me any feedback on it, but it is a chronic issue. We have to babysit that tunnel, and I don't know why.

We would probably use SolidFire more, except we're getting more bang for our buck with our purchases of ONTAP right now, and the deal we made with NetApp, so it's more of just a cost decision. Because we're going NAS, it doesn't really fit the pattern of where we're going, because everything is being presented via NFS, so it's just block storage. That would be the reason the footprint is not growing.

However, as long as the price is right, it is a no-brainer on block.

If they are using block storage, then it is very user-friendly. It's easy to use out-of-the-box. I was not a storage admin when I came to this team. I was a server guy, so it was all new to me, and SolidFire was the easiest thing for me to pick up. We had old 7-Mode systems. We had C-Mode. We had Isilon systems on EMC, and SolidFire was in a day, you knew how to do everything. It is just a real easy setup.

We don't have a reason to not use solid states. I don't know why we'd use anything else at this point other than solid state.

Most important criteria when selecting a vendor:

For a storage vendor, it is price and stability are probably the main thing. We like vendor support, but we have a huge internal IT shop with a lot of engineers, so we don't need that much support and hand-holding. It's really the following:

• Management
• A cost decision, who gives us the best deal.
• Stability.

If there's stability, and we haven't had stability issues with NetApp, they are a better deal than EMC, so that's why we've been using them. We were an EMC shop until three or four years ago.

We previously had another storage vendor, and we would recompose desktops of 350 VDI desktops or virtual desktops, and it would take us 10 to 12 hours. We then implemented the SolidFire on that same subset of users, the 350 desktops, and we could do it in an hour and a half. It's almost a ten-times savings as far as time for recomposing in our environment or infrastructure.

• The simplicity of it
• Ease of use
• The flash array
• Performance
• Reporting

We are looking for, potentially, on the Active IQ reporting side, to do reporting based on the datastore. Right now, I can report on the whole SolidFire, or I can report on just a certain datastore or a volume. I'd like to take all of my VDI infrastructure, which as an example would be multiple datastores, which I would select, and then do reporting on that for dedup, compression, IOPS - all the different metrics that we currently measure in the SolidFire. It'd be nice to be able to selectively pick what you want to monitor, as far as reporting.

Very good. With the exception of a couple drives that reported bad, we've had zero issues from that thing in two and a half years.

Scalability's easy. In the automotive industry, we don't know how fast we're going to go and when. It's really unpredictable. So, the scalability was a big feature for us when we were looking for a new storage vendor. We've already added to our cluster two separate times by adding nodes. We've already done that process twice in two and a half years.

Hit and miss, to be honest. I've had some bad experiences, I've had some good experiences. All-in-all it's been good, above average, but I have had some unfortunate experiences.

On the negative side of it, I'm working on support, and I would get not enough detail or the support tech would say, "Check the manual." They'd send me the manual, and its a 680-page manual. We're all busy. I have a job too. I don't have time to read through 680 pages. Send me the three, four, five pages that I need that's effective to my problem.

And ultimately just follow-up in another situation where we're working through a case. They tell me I'm fine. Then, a month later, they tell me I've got a problem on my SolidFire array, and all of a sudden. That was not told to me for the past 30 days, and it kind of caught me off guard from a communication standpoint. To me, the customer, I felt it could have definitely been improved.

Number one was reliability. We had a competitor of SolidFire that was never up, we had multiple downed outages where our whole business was down, and we have 20 dealerships or rooftops. Being down is not acceptable. Obviously, reliability was a big thing, and then, obviously, the scale out and getting to a flash array for VDI was very important for us.

In terms of solving those challenges, it's simple, it's straightforward, it literally just runs itself, and the scalability. When we need more space or storage array, we can just add to our cluster, which is huge, because we can't predict growth in our industry for automotive.

Very straightforward, very easy. We put in a four-node cluster in under two hours, three hours. It was very simple.

We were down between SolidFire before it was acquired by NetApp, so this would be even pre-merger, and our other one was Pure Storage. We chose this solution because of the flexibility to scale out compared to the competitors, such as Pure; along with cost, at almost about a three-to-one cost difference.

Operational costs, flexibility. The more nodes you add the more cost it is, but it's definitely significantly cheaper compared to other competitors that are on the market.

We did not want to consider hybrid storage because we previously had hybrid storage, and we had problems with our VDI, our virtual infrastructure, to where we wanted to get flash array. All flash was a big deal for us to get to.

Primary use case purchase for us was VDI or virtual desktop infrastructure, with the intention of VDI and our whole assortment of load, for the server infrastructure.

Important criteria when selecting a vendor would be reputation, accountability. Support was big for us as well, knowing that support will be there, ten-year, long term. And just durability, and knowing it'll be there.

Regarding advice for a colleague researching this type of solution, do your research, obviously. Check it out, give SolidFire a fair chance if that means you're demoing, bring it on for a bake off, definitely do so. It's worth your time and effort to look into SolidFire and what it has to offer.

It comes down to reliability, and it just runs. It just literally runs itself, which is all I could ask for.

We went from huge NetApp arrays to essentially a half a rack with the same amount of space that was required as far as data drives. With the footprint being smaller, and performance being way up, we're able to increase IOPS, which will give us better capability to actually mimic the production network on a government network.

• Power
• HVAC
• Density of drives

The square footage for doing development is at a premium when dealing with government networks. To be able to put a lot of IOPS in a lot of high-speed performing drives in a very small location which requires very little HVAC with very little power, it is very valuable to us.

With our Solidfire, we're going to be doing things like DevOps for ease of use. We're going to be able to expand in a condensed environment with a lot of IOPS to create a very small footprint on an all-flash array using a web interface, which makes it easier for some of the lower technicians to use it.

I'm seeing what I want to see. They're expanding and doubling the I/O per every 2U on their new 19210, or something like that. I'm looking forward to getting in there and testing it out as well. I'm really liking the performance of the network.

However, it would be good to provide administrative access at the root level to be able to do things with the system, if need be.

It seems pretty stable. At the beginning, they bring their engineer out. They actually set it up for you. That was a really good thing.

However, when we wanted to change how we want to do it and basically start it over, do some hands on training with some of the architects who would be working on it, one of the issues was there wasn't an admin password at the BIOS level to restart it over. We actually had to contact NetApp to come back out, or we had to contact them and get a one-time password. This was painful.

This thing uses a model like Lego blocks to be able to not only mix and match different models, but to be able to expand or scale out based on what the demand is currently. It gives me the opportunity to go from two to three nodes up to as many nodes as I need without having to put out an upfront cost that's way high. Especially when you're talking about development networks, this gives me the opportunity to provide the customer with a solution upfront which is high-performing. Then as the years go by, I can scale it out way further, especially if demand increases.

Tech support was fine. The fact that I had to contact tech support just to get administrative at the BIOS level was kind of painful. But other than that, yeah, lovely.

Density is the key here. Getting high-performance drives in a small package was what we were looking for. We still continue using NetApp, just a newer system, and when we want old flash arrays.

Definitely give it an opportunity; put in on a network, then put the different loads on the system and show how they do QoS across those loads to make sure you've got the guaranteed bandwidth.

Because I'm also a PC guy for my company and in Pakistan there is not a third-person specific guy for a client or specific job, we are also doing the very tough job as the IT specialists. So I have to look after the solution, the technical stuff, and also the deployment. I am personally working in all three departments. I have to, because it's my job as the head of IT in my company. We are a resellers, so we are actually giving solutions to our customers. Regarding SolidFire, it's a very good storage solution when you are looking for a software defined data center.

SolidFire provides seamless performance across your storage system when you need to scale up. Other solutions(software defined data center) do not provide that much of scalability.

If customer doesn't want to learn a lot of software stuff and only wants to learn one piece of software, and does not want to learn the storage system; that's where SolidFire comes in, because its software defined and it is really good for software defined data centers(SDC) and virtualization.

What I mean is, if a customer doesn't need a centralized storage system but does need a data center solution which is capable of being an agile software defined storage system, then they should choose SolidFire; but if they need a big centralized storage SAN, they shouldn't choose SolidFire.

For a customer who purchased SolidFire, they don't had reported any issue regarding it's performance or scalability.The end result is better productivity due to seamless performance of any application.

In terms of SolidFire's most valuable features, simplicity is the key component and key feature of SolidFire. It meant that the administrator or the user do not need to learn about storage RAID Groups or something like that, but they only need to provision the storage space that they need for the host. SolidFire's is being used in NetApp HCI(Hyper-converge Infrastructure)solutions that comes with a Deployment Engine to make the solution more simpler faster and easier to deploy.When you need a software defined storage system SolidFire's is really, really good.

When a customer need complete "Software defined Data-center Solution" with compute and storage; then NetApp HCI is far better choice than any other HCI Solutions.

SolidFire's really simple to deploy. you don't need to learn a lot. When you compare the NetApp storage system with SolidFire, both are very simple to deploy, but comparing to the other products from NetApp, SolidFire is even more simpler to deploy.

SolidFire should start from two nodes instead of the four nodes. That's the only thing. In a lot of solutions, we have to use four nodes, that's the better thing. But as a starting point, two is better. That's why their starting point is expensive. 

There is another thing - they should have a mixed/Hybrid disk option too; like other solutions have. If you get around the two things, then you can also compete with the cost. The others have mixed/Hybrid disk options. That's why they are cheaper.

Collectively; stability is good.

Scalability is really, really good with reference to other storage systems. Other software defined data centers do not have that much storage scalability. SolidFire is very scalable and that's really good.

The starting point of SolidFire is really a little bit on the expensive side, but from a scalability perspective, as a total cost of ownership, if you go for it then it's not so expensive.If you want a scalable solution, of course, the starting point will be expensive.

The product is really amazing, so I will not even think about the technical support because it's so easy to manage. There is no problem when you apply the solution according to the requirements and the documentation. You don't need support. In comparison other storage systems like Oracle, and others, have lots of problems and you need support. But with NetApp I only need the support to replace the component and that's easy to do. Also, the fault frequency of NetApp is really low compared to other storage systems.

I didn't experience any kind of problem which took a long time with NetApp support, but I did find a lot of problems which held me with support for other storage solutions like Oracle and EMC. This is very important because the backend engine is very strong. This is really a very smooth experience through the warranty period and of course after the warranty period. I got very good support from NetApp.

Initial setup is very simple as SolidFire build for SDC.

One thing I would advise people is regarding the price. This is a very important point here. When you go for being a gold partner, not as a customer, then you should get a better discount. And if you get a better discount from NetApp, then we can compete with any other solution. But here, it is really important to tell the customer what the total cost of ownership is, not just the starting point, because at the starting point, it's still a little bit expensive. But when you have a back-end gold partnership, then you can tell NetApp, "Okay, NetApp I am competing for the cheaper solution." Then you can get extra discounts to sell the product.

But then in parallel, you must inform the customer. If you want a Ferrari, you have to pay for that.You have to tell the customer, "Technically, it is very good." You have to convince your customer that if they want technical ability, agility, simplicity and real ease of management plus a lot of other things, then you have to pay for that. There's no free lunch at all. Somebody has to pay.

We are in a crisis situation for everyone, for customers, for partners, for principals, manufacturers. We need a specific solution to move on to the new normal. So from the available solutions, HCI (Hyper Converged Infrastructure) would be the future of IT. HCI is the future, it is easy faster to deploy collectively. You don't need to buy frequently - just buy once and scale out when you need more resources.

So, I think for the customers want their IT to have a continuous support and response for business needs, then they should choose HCI solutions.

I will advise to choose HCI from NetApp, others can choose HCI from other hardware manufacturers, but the future I believe is HCI.I believe and recommend that the best HCI is the NetApp HCI(The back-end Storage used for NetApp HCI is SolidFire)

The difference between SolidFire and other storage is that it is a stand alone storage. No one has that good storage software or that scalability and performance. On a scale of one to ten, I would rate SolidFire an eight.

The only thing is the initial cost because we have to convince the customer that he has to pay more. That's the only thing in this case. It's not an overall technical, scalability, or simplicity issue. Only the price factor brings it down to eight. Otherwise, if any NetApp partner is a gold partner and they get the good discount then I can easily give SolidFire a 10.

The solution is primarily used as an on-premises VMWare based application provisioning platform.

One of the most valuable thing aspect of the solution is the fact that it's all in one and all in a very small physical footprint. It has all of your major components, including your storage area network, servers, and networking footprint.

The delivery of the product is very fast and the solution itself deploys quickly, it is up and running within hours.

The product is competitively priced and technical support is good.

You can easily and effectively scale this solution. It's one of the main selling points and one of the features that makes it far superior to competitors.

We've been dealing with the solution for about three years now.

The stability of the solution is amazing. There aren't bugs or glitches. There isn't freezing or crashing. I find the solution reliable.

The solution is unbelievably scalable. It's one of the solution's main characteristics that differentiate it from other similar products. If a company is looking for something that can scale easily, this is that solution.

The technical support of the solution is quite good. They are responsive and knowledgeable.

The initial setup is pretty straightforward, deployment is lightning fast and very easy.

In House

The pricing of the solution is fair and definitely competes in the marketplace. We have no monthly cost. We bought the product outright with all licensing included and have support with them for three years.

I'd recommend the solution. Implementing it is a breeze, support is good and scaling is easy. If you do not have a lot of technical capability on-premise, that would certainly be a leading reason to look at this solution.

On-premises

The most valuable feature of this solution is its scalability. That is the biggest benefit of using this solution. 

The entry-level for this solution is so high that we had to use other solutions for some of our smaller office locations that are in different parts of the world. As a consequence, because we could not use it across our entire organization, we have changed to something else. I would like to see the entry-level changed so that you can do really small systems with SolidFire.

This solution would be improved if it were made to be more compatible with other products.

We have been using this solution for two years.

This is a stable solution.

Scalability is the biggest benefit of using this solution.

The technical support for this solution can be good and bad. If you have a really big issue then it's bad because it takes quite a long time for the issue to be solved.

There are different levels of service. You can have a technician who is able to do the troubleshooting and is allowed to set up the commands.

If you just have questions with no due date, or you just have to fix a small package, the support is good. You always get the answers you need. However, in critical situations, we have had problems. For example, in the last three years, we spend three or four hours on the phone with support where nobody could escalate our tickets. 

We have had to stop using this solution because the price was too high for global implementation. We are now using NetApp MetroCluster. We were happy with the functionality and switched only because of the price.

The initial setup of this solution is straightforward.

I would say that you can deploy this solution in an hour if you know how to do it.

The price of this solution is more expensive than others.

The suitability of this solution depends on the use case, so anybody who is researching this solution should take care to consider their use cases first.

I cannot think of any additional features this solution needs, but there is a long list of improvements.

I would rate this solution a seven out of ten.