I would like to have more administrative rights, for example, root-level administrative rights to the underlying OS of the storage array. We want more access to the kind of underlying infrastructure of the storage array rather than relying on support. However, most companies are looking to have more managed solutions which is the opposite direction of what I want. 

I don't find anything that is needed in terms of improvement. They are now gradually upgrading everything and are more powerful than we have.

It was a bit expensive.

HPE Nimble Storage's cost is very high, making it one of its downsides. HPE Nimble Storage's cost is very high compared with any HCI solutions, like HPE's SimpliVity, Nutanix, VMware, or Dell VxRail.

A person requires a special skill set to operate HPE Nimble Storage because there are hardly any solutions similar to Nimble Storage in the market.

The product's support can be made faster and more responsive if HPE trains its partners in such a way that they acquire more expertise on Nimble Storage through the training sessions.

HPE Nimble Storage is dependent on only VMware vSphere Hypervisor. In the future, HPE has to make Nimble Storage feasible with other hypervisors like Microsoft Hyper-V, Nutanix Acropolis, and RHEV, meaning it should be hypervisor agnostic.

Its pricing could be better. It's expensive.

In terms of features, it just works. It's in there, and it just keeps working. I don't spend a lot of time actively interacting with the Nimble environment, and there aren't any particular features that I want to see because it just works. I don't have any use cases where it's falling short.

Nimble Storage could increase its flexibility by adding more protocol options. Nimble mainly uses fibre channel protocols, whereas many other storage arrays support fibre channel, iCSI, and NFS protocols.

I want it to be an active-active array. Nimble would be great as an active-active array because then everything checks out. It would give a feeling of comfort.

Pure Storages have some option, which allows us the ability to directly download that host inventory and volume report. We can receive everything directly generated from the storage itself. We do not need to use some other web browser or something. If HPE Nimble Storage could do the same it would be a benefit.

While we're attaching the host and creating a volume. Afterward, it is a bit difficult compared to all other storage solutions. In other solutions, we can directly search for the host, but in Nimble, the drop-down is difficult.

As far as what future versions of the product should include, I don't have any experience with what it is lacking. It is really up to the technical guys to determine what is potentially missing from the solution.

You could argue that it would be preferable if everything were cheaper in order to save taxpayer money. While that would be nice, they are comparable to what else is available on the market and they are competitive.

Right now, all Nimble arrays offer data compression to disk, but only the Flash (SSD) arrays offer in-line deduplication. I’d like to see in-line deduplication extended to Nimble non-flash (called “Hybrid”) arrays, even if it’s only the C500 and higher controllers that support it.

I would like to see more integrations. They might already have them, but I want to integrate it with the different hardware we have.

When we’re setting up the solution, making options available regarding the replication tool mechanism would be ideal. There's a Nimble storage-based replication. They need to ensure that a customer beforehand understands what they want to do. When I joined, and I'm only about a year and four months in my new role here, I found that there was a gap in understanding of the type of replication they wanted and what they got. This is why right now, we're trying to change. Instead of having storage level replication, we're going to now buy Veeam so that we can run the applications active. That's how they want it set up. They didn't understand they could do that. They need to give options.

In terms of the setup, when someone is buying it, they should actually maybe ensure that they cover the admin portion. They should sit the user down to discuss and say, "If you want to deploy a solution like this, do it like that. If you want it like that, you need to buy this extra, maybe it's Veeam, and ensure you do it like this." That way, they don't just buy Nimble to set it up, and then it's not doing what it's doing. Then they'll think the solution doesn't work. Meanwhile, it's how they implemented it.

We are doing a hybrid and are moving some machines to Microsoft Azure to run in hybrid mode. We are checking the availability of extra software-defined storage so that we can configure it.

I wish they would put the InfoSight page back the way it was. I got in it for the first time about two years ago, and it looked the same for about a year and a half. Then, about six months ago, it changed. There are different options now. I can still get to where I need to go, but it feels like it takes longer, where before it didn't. Also, I felt like I had a lot more options before. I have to do a lot more to digging now to get to where I need to go. I just wish they had their old page back.

The solution that I have is a hybrid, not a full flash. The hybrid version could be improved.

The only thing that I can really compare Nimble to is all-flash because, right now, Nimble is a hybrid solution. I would like to see them come out with an all-flash alternative.

I'd also like to see them incorporate tools that let me get granular with the VMs. I want to see an individual VM, I want to Snapshot and recover an individual VM. Those are the kind of daily operations features that I'd like to see.

The solution could reduce its price.

I don't think it is officially released yet, but the main reason that we chose Nimble is because of the sync rep feature. So, I would like to see that further evolve. This feature will be essential for our setups.

I would like to have integration into cloud providers, apart from HPE.

The release cycle for the firmware upgrades could be improved. They are a bit long.

They only use block-level storage and with NetApp, we were able to use file-level storage. That means we could leverage NFS or we could leverage SIFS and we could actually host a file server on the storage solution itself. We could do that with our NetApp solution. Right now, they only use block-level solutions, so you can only use iSCSI.

The product is great.  When sizing the array based on requirements the option to add more network cards for throughput would be something to help clients.  Product Support is best in class as always.

The solution’s stability could be better. The tool’s pricing is high and depends on the partner.

I would like to see some additional features included in the next release.

I would like to see the network portion of the product improve, especially with some of the things which are coming out from Aruba and HPE. Both are innovating more of an automated networking. I would like to see our Nimble meld into that and do some automated networking.

The solution requires a higher availability.

The pricing of the solution isn't ideal. They should work to make it more affordable. It's very expensive.

I'd like to be able to configure the solution from vCenter, which isn't possible right now.

It would be great if the solution offered even more integrations and plugins.

I would like to see more advanced analytics in the next release of this solution. Perhaps it will help us to see things that we're not seeing already.

We are fairly happy with the product. I think most of it is definitely satisfying the requirements. But one area, for which we presented our feedback to Nimble, was how we can centralize, i.e., monitor the multiple arrays in a centralized manner, from a single viewer/ single console. Let's say, for example, you buy more and more form Nimble; how can you view all of them in a single console, i.e., viewing it in a single window-pane, such as viewing the health checks, etc. This they have taken up and I think they're going to get back to us, in a month's time or so, is to provide a solution.

The storage can be used in two modes, i.e., in a standalone mode or in clustered mode. In case you are using it in a clustered mode, they are integrated together logically and hence, it is possible. However, in case I'm using it in a standalone mode, they have more than one array, then instead of logging into every single array, how can I get a console from Nimble to view all of them in a single console, i.e. when they are configured in the standalone mode?

The only thing I'm really looking for in my next array is some hyperconverged, so if they had something in that space... But I know they have SimpliVity so that is probably not going to happen.

I really would like to see synchronous replication. This is something that when we have multiple arrays in our environment and being able to do something like a zero RPO. Being a law firm, we really want our data to be protected all the time. Although, we are really rocking the Nimble snapshots and you'll hear them talk about the maximum snapshots that you can have; there's no performance impact and it's just really cool to have that kind of data protection. We're utilizing it but getting further, so as to get the knob all the way down to zero RPO, that would be a really cool feature. I hear that they're working on it.

Honestly, the only thing that could really get better is the price. Although, the price on these units isn’t too bad at all.

I think the scalability of HPE Nimble Storage could be improved.

In Venezuela, we have to purchase the solution for two years and cannot obtain a secondary storage platform. So from my perspective, the scalability is not as easy as that of 3PAR StoreServ or HPE Primera.

I would like deduplication by default on all the volumes. I still don't understand (or know why) dedupe isn't enabled for the templates out-of-the-box. We have to go in and manually enable it each and every time. It might help to have a software revision for something like this, or even a more intuitive local dashboard, not InfoSight, but Nimble's local dash. That would be great.

The UI and local dash need improvement. They have changed it around a little bit between firmware upgrades. When you do that, things are out of place and not where they used to be. There is also some granular nitty-gritty analytics that we would like to get down into deeper, like when you view volumes, you have percent-free space, and the ability to sort those volumes by the percentage or sort by the amount of data: the megabytes, gigabytes, etc. That is no longer available in the new UI. I would love to see that improved, where you can sort down the list. 

While InfoSight is very useful, if you widen your scope of range when you're trying to look at a month of reports, the averaging that it does sort of skews things a bit. If you drill down further into a day or hour, it looks a lot better. I wish they would have a chart or GUI that didn't average as heavily, just giving you minute spikes and things that you could see on a wider view.

Actually they already have a solution for all flash. So, we are planning to use that in our deployment. So, actually there were no issues with features, and we are waiting for the upgrade.

It would be good to have a built in way to index snapshots so that we’d be able to granularly see what files were in a snapshot and restore at a file level. We’re using an add on product from Commvault to achieve this.

An area that needs improvement is extending the life of the device after five years.

What I have seen so far has been sufficient for our needs.

My version does not have dedupe, it has compression. I know that they have dedupe now. I just need to buy it. 

I would like to see native network attached storage (NAS) functionality. Our customers are looking for NAS, and Nimble can't give it to them.

There are customers who want to do some different things with the Microsoft Resilient File System. There are some customers who want to do different types of connectivity. I do not know if I would call that an improvement, necessarily, because if you want that, you should get a different product.

Currently there's no deduplication, and I would like to see that. Also, it's still slower than flash storage.

I want to see the full integration with OneView. I know they have started it, but I haven't had a chance to look at and evaluate it.

They keep talking about doing one-to-many replication. This is what we'd like to see, just to help with data replication.

This product could improve by adding support for file protocols such as CIFS or NFS, the ability to seamlessly migrate from one volume to another, and the capability to perform in-line deduplication.

Volume encryption must be done prior to occupying data on the volume. A data migration is required if the encryption requirement changes. This could be improved by being able to turn on encryption on previously created volumes without the need for data migration.

There is not a lot of room for improvement. All the items that we wanted to have implemented when we first started using Nimble arrays were either already on their road map, or added after listening to us, and all bar one have now been fully implemented.

These included:

• Active Directory authentication for logging into the management of the array.
• Full audit logging of all actions taken on the array.
• Full VMware implementation for managing the arrays. This has and will continue to save us a lot of time.

Over a year ago, Nimble was available with either iSCSI connectivity or Fiber. Controllers can be upgraded or replaced with either of the connectivity interfaces. But, a combination of iSCSI and Fiber unit will provide flexibility in connecting SAN with existing network switches or fiber switches which can reduce overall cost and help with connecting high IO devices to the fiber interface.

• Improved snapshot overview in array management GUI.
• Possibility to apply centralized policies across multiple systems.

We would like there to be more enhanced features.

It would have been nice if the vendor had informed us of a known bug that occurred during the firmware upgrade because we could have prevented our only major failure.

I would like to see greater integration with Microsoft's Hyper-V platform. We are one hundred percent Hyper-V and no longer have use for the expensive VMware platform.

Something I would like to see is more clarity regarding the positioning of Nimble versus 3PAR. I am struggling with the boundaries within which these two are competing.

I would like to see SSL Certification. For someone like me who has been in the field for a long time, it’s easy to overlook. However, for someone who is fresh out of college, it will be hard for them.

While I was not present for the sizing discussions, I believe there was some lack of discovery in selecting the model and features for my client. Even though the flash size on the array was doubled from Nimble's initial recommendation, the SAN still has latency issues during the largest dataset imports, due to CPU limitations on the SAN model that presales had recommended for this client. Certainly not the end of the world, and it can be fixed with an upgrade, but it was somewhat frustrating for my clients. I suspect estimations were used as to what the expected load would be, without a full understanding of the data processing being performed.

I used to have some gripes with how the GUI worked such as folder organization but the developers listened and released that in a new firmware version. I do want close to real time performance metrics in InfoSight.

Support for near synchronous replication.

Currently the smallest snapshot window is 15 minutes, that means if we replicate data from one array to another array we will be up to 15 minutes out of date, this is the RPO.

We have a small number of systems that need to be replicated in such a way that we do not lose any data if we need to fail-over to the replica on the second SAN, or at least it should be only seconds behind.

Some SANs, like NetApp, have a synchronous replication feature, so when data is written to SAN1 a copy is sent to SAN2, only when this has been written successfully at SAN2 does the data on SAN1 get flagged as written. This means both SANs are in-step.

This is great if you have the SANs locally and have a fast connection, but if they are in different data centres and the WAN link is slow you get latency issues while the data is written and acknowledged by SAN2.

When I talk about near synchronous replication I am thinking of how DoubleTake from Vision Solutions works, I have been using this for over 15 years to do this type of replication.

This works by copying the block of data on SAN1 directly to the local volume, but taking a copy and putting it into a buffer. This means latency at SAN1 is not compromised.

The copied block is then sent across the WAN to the second SAN, where it is written and acknowledged back to the DoubleTake system on SAN1. Once the acknowledgement is back the local copy on SAN1 is deleted.

This means if any latency on the WAN does not impact the speed of storage on SAN1, although the data on SAN2 could be a little bit behind, but in most cases we are talking seconds at the mots, and once any congestion on then WAN link has gone the replication is nearly as fast as a pure synchronous system will achieve, so for most purposes it is as close to real-time as required.

If the WAN goes down, or you need to take SAN2 off-line for any reason, the only impact on SAN1 is the buffer grows with data to be replicated, and once the WAN or SAN2 is back on line this is flushed to SAN2 as fast as the WAN can transfer the data.

You need a ‘re-synch’ feature in case the volume on SAN2 is corrupted, this checks the CRC on the data blocks between the two systems and re-transmits any blocks that do not match.

There should be faster interfaces.

It would be nice if you could use something other than the proprietary rights on it.

It would be ideal if all these enterprise-class high-performance products would come at a cheaper price.

The initial setup could be simplified a bit. 

The solution could improve by having more granularity. For example, having the ability to go deeper into specific IO channels or specific latency issues that can happen would be a benefit. HPE has this ability on their side but it would be useful to have it on our side.

Replication reporting is a bit difficult to figure out and could improve in the future. It would be helpful if they had a breakdown detailed report that one could provide to upper management.

From what I've seen personally, customers that have Nimble don't really switch away from it. When newer versions come out, they usually have everything in terms of all the support and all the additional components that might be needed to help.

With the newest models, you're getting decompression and dedupe on some of the entry-level models, and this fixes a lot of issues for customers. Therefore, I am not hearing anything at this point. They just released the newest version, and so far, it's going well.

I haven't thought about what I might want to see in a future release mainly because we're trying to figure out where we want to head next: if we want to go with the Synergy system or if we want to continue down the hyperconverged path.

I think HPE is headed in the right direction. From what I know, their management is actually putting Nimble on the forefront to be the leader, and I think that's where it belongs. To stay ahead of the game, they need to put Nimble in front and stay there, and ride that horse throughout.

For me, this product has everything I need in there. What could be improved is performance monitoring, i.e., if it can get even better than what it is now, I will be happy. For example, there should be an exact drilling down to any workload/any virtual machine/any identified issue. So, if an issue appears and there is latency, there could be a faster or easier way to know what the reason is and how we can resolve it with Nimble.

They should improve the solution's pricing.

The problem is the price. It needs to be improved.

I would like to see an added feature to auto-fix, or a dynamic alerting system on storage. This is very important because we would like to prevent a disk failure before it happens.

If we had some sort of AI in place to alert us then we could replace the disk before it occurs.

Also, we would like to receive alerts if space is over the limits. This is necessary for us.

The amount of firmware updates released seem excessive.

The admin GUI is pretty ugly. It works, but it requires a plug-in. I never had to use it since the VMware vCenter plug-in is so good.

Just continue the updates of technology features; I expect there will be more updates in capacity and IOPS as models upgrade in the future.

Small tweaks to the management UI for usability.

No areas require improvement at this time. The arrays run very well and updates are flawless when completed.

For me, Nimble has two main problems.

There is no active-active controller, which means that we can only have one controller online at a time. Replacing the controller is what I see as the only major issue, although I'm not sure that HPE can do this.

Nimble has a limit for objects. We have it configured for VMware, so if you have a laptop machine then you have a problem because of this limit. Also, if you have a virtual desktop with a lot of VMs, such as 2,000 to 3,000, then it's a problem because the window has a limit.

More reporting is probably the only thing that is really lacking. It would be helpful to go to the business and say, "This is how we've evolved with our solution, and this is why we need more." Being able to put forward a business case with data to back it up, essentially.

It's purely as and when we get to the point where, we're either on a refresh, or we just need more capacity, or performance, then it's something to take to the business as proof.

I'd like to see more granular quality of service rules. So things like: I think currently there's not much room for maximum IOPS, but there's not an option for minimum IOPS for a given volume.

It's less about giving us more features and more about giving us ways to contractually guarantee the features that are already there. So something like performance is the classic one. It's more valuable to me to be able to contract the performance that's there, rather than have a new way of doing things, because customers are not interested in signing up for the "best effort, maybe" services anymore.

I would like to see integration with OneView, at least for reporting or some sort of monitoring. Basic stuff like that would be nice. I can't think of any area that is lacking for what we use it for.

All-flash arrays.

Admin interface can be slow.

There have been a couple of instances where we discovered previously unreported bugs.

I'd love to be able to put larger spindles into the unit, but I get that that is how they make their money, by selling you a larger unit.

Competitors, such as Dell EMC, make use of NVMe storage. They also use the STM storage module. Primera only has a maximum of eight. In contrast, HPE Nimble Storage does not use NVMe and this makes it challenging for us to convince the customer, who is sometimes aware of this technology, to go with it.

HPE does not have sufficient storage. Overall, the customers in need of the solution have sufficient network data storage. There is also SAN storage in one solution, in a single product. I have made three requests that the storage be capable of being samples in a single bundle. We have the protocol as an example when it comes to the product. Yet, if the customer needs to configure NAS storage, this necessitates the purchase of separate software. As the competitors have already launched the solution with NVMe or STM storage, we have also done so with Primera. The same holds true in respect of Dell EMC. In Indonesia we encounter customers who are knowledgeable about the storage and infrastructure and enquire about the solution as it concerns HPE storage. 

So far, the only area I have noticed that requires improvement is the ability to console directly to the device. The device is supplied with a special adapter that requires a female-to-female serial cable. If standard cables (preferably USB) cannot be used, then complete solutions should be provided with the device.

HPE Nimble Storage could be improved with some critical application or servers. Currently, we have to compensate for this by using flash storage. In the next release, HPE Nimble Storage should include active clusters and a hybrid file service. 

Nimble is costlier than EqualLogic, for example. If they could reduce the cost, that would be ideal.

Third-party integrations: A lot of products tend to come out and support NetApp, EMC, and various other storage managers. Nimble tends to be lagging behind in that space, and it's not completely their fault, but I think they could do a better job pushing for those integrations with various third-party products.

The backup solution that they are using now probably has not been deduped. I think this is probably sort of one area that the customer is a little bit concerned about. It is a timing issue for them, though it is still working at the moment. It is just not giving them enough header space that they thought they would have because the dedupe does not seem to be working as well as it could. So, it is something that needs to be reviewed and to make sure the customer has not clicked a certain feature. 

I know for a fact that once they get in touch with HPE or Nimble, they will obviously look at the box remotely, see where there are potentially fine tunings, and if it is just a question of the customer having time to allow them to come online and review it, hopefully it will get solved.

Apart from that, everything else has been positive.

It’s just perfect the way it is!

Nonetheless, it would be nice to have:

• I need a real API (REST), and Nimble Storage made a REST for me (and all the other customers), so the product is perfect for me! Most of the other roadmap features do not interest me.
• It would be nice to have a ”virtual Nimble Storage box” in the cloud, so that I can replicate data to the cloud (and clone VMs and spin them up). This is in the roadmap.
• I also want replicated snapshots on CHEAP space in the cloud (like Amazon Glacier).

Perhaps more stability to the InfoSight VMVision feature. A couple of times last year, this feature had stop working for us.

Well, Nimble Storage have had some difficulties/interference during this year with their predictive analytics service InfoSight. Read more about it here: https://www.nimblestorage.com/...

(almost) every Nimble array in the world reports back to InfoSight with data and statistics about the VM’s and connected hosts, which InfoSight automatically analyses and presenting it in a neat way for the user. A very good service and a matter affect, recently (the last couple of weeks) it have worked flawlessly, and hopefully this will continue.

I haven't come across any so far.

The cost of this storage solution should be lower.

In the future, I would like to see a lower-end model that has Peer Persistence functionality.

The only area of the product that could be improved is if you have to buy additional storage, you have to buy a block. You can't buy two or four hard drives as you can buy for every other system. You have to buy a block of 24 disks, and that's normally very expensive. So if a customer calls and they want to expand the system for a few terabytes, you have to sell them an expensive solution, and it's the same for us. But that's just part of the system. That's part of the architecture.

A feature that would be a nice addition to the next release would be a filer option. A filer option so that you could connect the sim or NFS or chips like NetApp does for NAS functionality.

I would rate the solution a nine out of ten. We are very happy with the product.

The scalability could be better.

There is a new version of the Nimble and 3PAR systems called Alletra and they have a lot of new features. We have to check and have to give them a read, however, at the moment, there are some new features with cloud connections. It's a feature that they haven't finished yet. There's the possibility to move the workload to the cloud and back. At the moment, they don't really need any improvements.

When I'm competing against someone, I would like Nimble to be an active-active controller. As it is now, Nimble is an active-passive controller. If the customer is looking for an active-active controller, then we can't use Nimble and have to go with Primera. Nimble gives us a lot of good things, but without the active-active controller, it's pointless.

The stability can be improved.

The way the market is going, supporting NVMe storage would be useful as a step up.

The InfoSight continues to improve, but could use some more innovation.

The solution should allow for easier configurations.

We want to start storing data in the cloud really soon and we are looking at other options. If there is a solution that could combine Azure backup and interconnect it with storage on-premises and on the cloud, this would be great. This solution could already exist by HP, I have not done the research yet or know what the process looks like.

• Web GUI should be HTML5.
• Use smaller form factor disk to provide more spindles. Streaming Backup I/O overnight would dent performance if we were a 24*7 business. Most of our workload is 07:30 -18:30.

The quality of technical support depends on which member of the support team you speak with.

The product is near perfect, but there is always room for further development; maybe, more automation in snapshot management.

Easier retrieval of snapshotted data and more granular control is desired.

This is in comparison to something like Veeam or MS shadow copy, where restoration of files is easier in their respective proprietary GUIs.

Snapshotting with Nimble can be very granular, but the restoration from a snapshot is not as easy as restoring with other systems.

Scalability, in terms of being able to scale out, is not easy and should be improved.

In the future, I would like to see the vendor upgrade the hardware.

This is a tricky one, really. It does everything we would want from a SAN. It doesn't make the drinks, but all we're looking for is performance and reliability, and it does that.

Nimble offers InfoSight, which in turn offers VMVision. This is a terrific feature. You can immediately see which VM causes the highest IOPS and which has the most latency. The downside of this feature is that you can only see it on InfoSight and the data is always a few hours old. It would be awesome if this was in the dashboard of the Nimble and shown as live data.

Also, Nimble requires its DNS to work in order to interface with VMware vCenter is a little thorn in my eye, but I am being nitpicky here.

We heard from the Nimble pre-sales in 2015 that an inline deduplication feature is forthcoming (2016). What I know now is that this is not the case, dedup is only for all-flash array only, something to do with computational constraint. Well, we are a little unhappy with this news but because almost all the competition (VMware VSAN etc.) is also doing the same thing (inline dedup only on all-flash) I guess it’s not yet feasable with current technology.

The HPE Nimble Storage could have better integration with monitoring and machine learning system information solutions.

The dashboard can be improved. I would like to see more details on the dashboard in the next release.

Pricing.

I have a long list that I have shared with Nimble development team and am working with them on this.

The most difficult part about Nimble was the fact that it didn't have a standard length. It had a prolonged length to standard servers, which made it slightly challenging to get it installed and we had to switch racks around.

It would be beneficial for Nimble to have an NVVM and/or an all-flash solution.

Currently Nimble’s All Flash solution is SSD disk based. Nimble should investigate in having their All Flash Solution with NVM Ram (similar to HP) or memory storage similar to Violin Storage.

Scalability, see below.

Let's see, well you already have the old flash, I felt that was missing but, I don't know. It's a very good solution but the price is not the best, probably better pricing would be good but yes, it's a good solution.

Further improvements on existing features, always.