Dell XtremIO Reviews

VDI is one of the top mission-critical things we offer our users. This storage runs our whole VDI environment and barely shows a blip on I/O. Previously, we had ran the VDI on non-flash storage and when Windows updates came out, we had to install them in schedule segments so as not to overload the storage. With this storage, we do them all at the same time and there is no impact to performance if 1 or 100 VMs reboot at the same time.

Dedupe, compression and high I/O are the most valuable features. It is great for applications like Microsoft Exchange, ERP, SQL and VDI; basically saved the VDI buy-in from users, as now performance was seamless in comparison to a physical PC.

Get rid of the Java aspect of the GUI console. Basically, the GUI to administrator the array uses Java as its base to run on. Java at best is buggy and prone to loading issues, so moving away from this platform would be nice.

We had no issues with stability.

We had no issues with scalability.

I would give technical support 9 out of 10. Nothing is perfect but they sure are close to it.

We previously used deployed EMC VNX storage (and still use it for our lower performance applications) and before that, we had Dell EqualLogic. We switched to an all-flash array as we wanted high performance storage for our three most critical applications (Exchange, ERP and VDI). We wanted to do a full VDI platform for all our users and locations. We wanted the best experience for them, as any hiccup would mean a lower buy-in rate from them. This storage made that task much easier.

We bought it through VCE, so they included setup with it. Things went smoothly. When we did receive the storage, within a day or two, we had a controller failure but since it had two controllers, there was no impact to users. Support was fantastic and got it replaced over the weekend, and we didn’t even have to do anything other than authorize them into our data center to replace the failed part.

It is costly but worth it. If the network or infrastructure you have is always a sticking point to users or management, spending the bucks on an all-flash array can help win them over.

We looked at more EMC VNX storage but at that time, we were not aware of this offering. When we started doing the talks with EMC, our rep pointed at this product line and once we saw a demo, we were sold. After more research, it didn’t take us long to get the paperwork in place. We also didn’t look at other vendors, as we utilize VCE as our main infrastructure at our data center so regardless of what model or product line of EMC we bought, VCE would handle the support. This was one of the main reasons of going with VCE, so we wanted to carry it on with the new storage.

I wish we bought double the capacity but we only had so much to spend, as I would put every application/server on this array.

We use it together with VPLEX, which virtualizes the storage array with all its benefits.

This virtualization layer adds to the latency. With XtremIO behind the VPLEX, the response times are far below the response time we have on our other storage arrays, even with the SSD onboard.

The data reduction (deduplication and compression) is the most valuable feature in our business case.

We calculated a reduction ratio of 3:1 to get a positive case, and we actually reached a little bit higher (3,1:1). This makes our business case even better.

Even with this feature, the response time is far below what we received with our other storage arrays.

Another valuable feature is the guaranteed sub-millisecond response time for a 4K block.

It has no storage replication. The replication is done through the VPLEX. In some cases where we don’t need the flexibility of the virtualization layer, we could free up resources on the VPLEX by using the storage replication.

Until now, we have not encountered any issues with stability.

We did not immediately have scaling issues. Scaling up is, in fact, very easy. Just “buy” an X-brick of 40TB and plug it in. The system does the rebalancing automatically. Since we use a VPLEX, the scaling limitation lies with the VPLEX.

Technical support is good. The installation went smoothly from DEL EMC’s site. We did not encounter real technical issues yet, but the questions we had were all answered within an acceptable time frame.

Part replacements are done transparently without any intervention from our site.

We used HPE and EMC storage arrays, but the main reason we switched was the positive business case. We have a lot more flexibility (VPLEX), reduction of cost and floor space (XtremIO), due to deduplication and compression.

The initial setup of the XtermIO was very straightforward in combination with VPLEX. The setup of the VPLEX was little bit more complex, but XtremIO just needed to be connected to the VPLEX.

XtremIO is pretty straightforward about pricing. However, you need to look at your data so you can estimate, with the advice of DEL EMC, what data reduction ratio you will reach. In our case, a 3:1 reduction ration gave us a positive case compared to other storage arrays.

The XtremIO by itself without a virtualization layer has some drawbacks, like storage replication. I really would recommend them to install it in combination with a storage virtualization layer.

Compression and de-duplication are the most valuable features, especially when compared to other All-flash arrays like Solidfire, Violin Memory or Pure Storage.

Overall space usage has decreased, plus input/output operations per second (IOPS). Performance has increased exponentially.

The replication and DR capabilities could be improved, since there is no native replication technique with XtremIO. EMC wants customers to rely on EMC RecoverPoint for those needs or VPLEX for DR sorts. Native replication is needed.

I have used it for 3 years.

I did not encounter any stability issues, it is very user friendly and easy to use.

Scalability was non-hazardous and even online.

Technical support was exemplary with a very fast response time by Dell EMC.

We used to employ high-end and mid-range products from multiple vendors like EMC and NetApp before we switched to XtremIO for good performance and space economical-ability.

Initial setup of this product following stacking and racking servers is straightforward. The product is not complex in its modus-operandi and you have the bonus of sources and support availability.

Pricing is definitely on the higher side and licensing plus maintenance can come as a package. I recommend this as a good negotiation technique with EMC.

We first evaluated and selected VIMM (NVRAM chip) arrays which were faster than XtremIO but costlier. We then looked at new players like Nutanix, and even ScaleIO. But in the end, XtremIO turned out to be cost-effective and at the same time worth spending for the change.

Be it XtremIO or anything else, you need to see that requirements are aligned with the following:

- Data Centre space and cost.
- Storage growth trend and capacity forecast.
- Application performance and requirements.
- Operational readiness and implementation timelines.

We used the product for our internal customers (Banking). The performance really improved for the end users with reduced costs.

It has very good performance for an application which needs lower latency and a better response, for example, in microseconds. It's very easy to use in terms of administrative tasks on the GUI.

At times, it's difficult to track down allocated resources as compared to other solutions, like VMAX and VNX, because it's completely software-based.

The GUI could be modified more in terms of how the different components are linked to each other.

Yes, we had issues with data corruption on one of the X-Bricks. The issue propagated to other X-Bricks as well.

No, in terms of scalability, we didn't encounter issues. It was very easy to add X-Bricks to the system.

Technical support was great in terms of first response and further troubleshooting assistance.

Yes, we were using a VMAX solution before. This product was incorporated in order to get better performance at a reduced cost for our internal customers.

I was not involved with the initial setup phase. My role was more focused towards administrative and implementation tasks after the product was installed.

Again, both the pricing and licensing were handled by another group. So, I can't comment on it.

Yes, there were multiple options evaluated and also incorporated for some percentage of the workload, like NetApp C-mode array. Specifically, XtremIO was considered for Xtreme Performance as advertised by vendors at a lower cost as compared to existing storage arrays, like VMAX and VNX.

It's very good in terms of performance and the ease of use of its GUI. However, we had issues at the data corruption level, which resulted in multiple X-Brick failures. It's a big question for the stability and availability when we have enterprise level customers on such an Xtreme Performance array.

What I like about this solution, is that it is really fast and it's really good in compression. So you can put a lot of data in line on it. I've used it for cases like VDI in a healthcare environment because we use multiple copies of it. It is very good and has very good performance. It can handle big workloads. 

What needs to improve is that the end compression uses a lot of CPU, which makes it difficult or impossible to upgrade. I believe that it has improved in the upgraded versions, which we don't have yet. Also, the price and interface should be simplified too. 

The program is very stable.

We had some difficulty when we tried to expand our XtremIO because it had to be done in a certain way and only EMC personnel were allowed to do it. That wasn't very flexible. But it seems to be better in the new version. Scalability is always a problem. 

The technical team is good.

We previously used another EMC program, but it was not flash only, so that's why we switched to this solution. It possible to buy an all-flash system and, based on the compression features, we changed to this unit. 

We were not allowed to do the initial setup ourselves. So it was done by EMC personnel and maybe that could be improved so that our partner, our supplier, could do it themselves so that you don't need an EMC rep to perform the installation.

My advice to others would be to test it first to see the expected ratios. Do a check to see if they deliver what they promise. I would rate this solution a nine out of ten. To make it a perfect ten, I would like to see better scalability. 

I use XtremIO for backup and recovery, to allow connected applications to access storage, and to locate storage resources for specific users.

I've been using XtremIO for four years.

XtremIO is very stable.

XtremIO is scalable - it has expansion charges so that you can extend your storage size.

Dell's technical support could be better.

The initial setup was straightforward.

We used a vendor team.

I would recommend XtremIO to others and give it a rating of seven out of ten.

• The increase in available I/O dramatically changed the responsiveness of our Windows environment.
• SQL server limitations of tempDB and overrun were non-existent.

• Deduplication
• High I/O capacity
• Thin storage allocation
• Snapshot cost and speed

• Management: At the time, there was no snapshot scheduler, so I had to write XSnapCourier to address it. The sad thing is that even after the newest release, which includes a native scheduler, most customers using XSnapCourier chose to stick with it due to a more feature-rich experience.
• Planning: The VAR we went through told us we would get something like an 8 x 1 ratio between compression and inline deduplication after migration from NetApp. For this reason, we only elected for 20 TB raw. What we actually got, however, was little over 2 to 1, which was actually worse than what NetApp afforded us.
• Support: Even though we paid for a four-hour turnaround support contract, we would have to wait up to two weeks for a response from EMC XtremIO support, because they didn’t actually have the support staff to handle new volume.

We have had major stability issues. Over the course of two years, three of the four storage controllers failed. It took us over two weeks to get a replacement for the first.

Technical support was the worst possible. Regardless of the fact that we paid for a four-hour turnaround, we were waiting two weeks for support calls. When we did get support calls, the engineers were up to four hours late to the datacenter. After the engineers would finally show up, we would still wait weeks for parts.

We previously used a large NetApp array, but the issue was storage density. We were at the point where we couldn’t add any further disk shelves to the controllers.

Don’t buy this array. You’re paying for loads of magic beans, since it’s mediocre at best for a platform in a rapidly growing field. Look instead at Pure Storage or something with variable block deduplication. You’ll end up spending less and getting a better product with actual support.

We were not given the opportunity to evaluate alternatives. Upper management made the decision without the input of the engineers.

Don’t consider it. Look at a platform that has actual support. EMC is a big name, but their support model is terrible with an even worse model for implementation. For a platform you literally can’t touch without them, you’re stranded on a desert island with no help in sight.

It allowed us to shrink our storage footprint in the datacenter and it increased our performance.

Performance and deduplication. This is a very robust block storage option that offers both performance and data optimization.

It is great when a product can deliver high-end performance capabilities while offering a very competitive price point.

• The new UI could be simplified and the implementation should be simplified.
• The UI is easier to use than other EMC offerings, but still takes a pure storage admin to manage and maintain.
• Ease of use is key in the converged and hyper-converged world that requires administrators to have both hypervisor and storage skills.
• Arrays should be “set it and forget it”, including the replication features, which should not require separate products.
• Implementation and expansion require EMC or a partner to implement. However, to be a true scale-out solution, it should be something that customers can do themselves.

We have not had any stability issues.

There is a limit to scalability, but we have not reached it yet.

Technical support is excellent.

We did have a previous solution. This allowed us to move to a smaller footprint with better performance.

The initial setup was very complex. Have the manufacturer or a partner install it.

The pricing is simple but requires other products for robust data recovery and replication.

Make sure you understand the shutdown and redundancy for this product. Test these features to protect the storage controller’s journaling.