What needs improvement with StarWind HyperConverged Appliance?

There was a long delay with shipping, which we weren't expecting, but we were kept updated throughout. It wasn't made clear to us that we needed to provide our own SFP modules to connect to our infrastructure, so that could have been a bit clearer. It would be amazing if there was, perhaps, a more automated process for updating the hosts themselves, although the maintenance mode works well. Our StarWind command center doesn't seem to have all the options available currently, which support is investigating for us.

I honestly don't have anything to complain about. Nothing is perfect, though, over the past three years, there were two instances where our cluster had an issue, yet StarWind proactively notified me of those issues and effortlessly guided me through the process of correcting the issue with minimal effort on my part. Best of all, there was zero downtime to any of our VMs. The reliability of their product combined with the proactive support means we've basically had almost three years of smooth sailing and 100% uptime.

The only area that surprised me had to do with the fact that each host runs a StarWind virtual machine used to maintain the health of the vSAN, which individually uses 24GB of memory per host. You should keep this in mind when determining the resources you will need when selecting hardware and components for your specific setup. If I knew that, I would have chosen to add more memory per host to make up for the use. I had configured my setup thinking I would have that much memory available to use to provision additional virtual machines.

Maybe they could have a support portal that you can have direct access to a current support representative. During setup we had a bit of a tricky situation that had to do with switching. The StarWind engineer recognized the possible problem with migration before it happened. During test migration when things went down it would have been nice to be able to reach out directly rather then send an email to support and wait for a reply. It was not a long wait, however, it can feel like forever when in a tough situation.

There is not much that could be improved. StarWind was affected by the delivery problems typical for the market and could not be improved by different partners. What would have to be done urgently, however, is the adaptation of the hardware to the configuration. We could only decide on a model that was in the upper premium segment, although we only needed standard specifications. Therefore, we have a solution that is a bit too performant concerning our requirements. That said, this also allows us to try something out. The price was excellent, so we accepted it.

We need to be very cautious in following every step when updating the physical host. We must move over each VM and drain roles from the servers to ensure everything goes smoothly without interruptions. If this were a more automated process, this would be less taxing each time an upgrade is needed. That being said, even though there are some steps to the process, each upgrade has gone well without any issues. It just takes some notes and processes to get comfortable with moving things around.

I'd like a better UI and some limitations on "breaking it." Best practices ask you not to reboot both servers simultaneously, as it breaks the "heartbeat" between the servers. I didn't know that during my first week having it, and sure enough, I had some issues. Maybe there could be some notifications like "are you sure you want to reboot" or whatnot to prevent issues like the one I experienced. StarWind did release a VM that had some integration into the Windows cluster side of things. However, I found it easier to just use the Windows tools. Maybe if it was more robust, I'd use it. That's not the case at this point in time.

At times there has been a language barrier when communicating issues. That said, there was nothing that couldn't be worked through. This made it tough when trying to understand or explain some of the issues we were encountering. The user interface for the StarWind product could sometimes be tough to navigate and understand. However, that was not a deal breaker. We would like to see that simplified in the future. Another feature we would love to see is a dashboard that gives a comprehensive view of our entire StarWind infrastructure along with metrics on how everything is running.

The user interface needs an update for the Windows client. I would prefer it if it was all web-based. This way, you do not have to manage an installation program on a server. If it was web-based, it would work on any platform. With the new interface, maybe some more reporting features would be nice. Also, some notifications would be helpful for issues coming about with the system. We would like to be able to shrink volumes. We know we can increase volumes when we need it, however, shrinking would be helpful.

We haven't yet really found anything that we think needs immediate improvement. I am clutching at straws to find something if I am honest so would probably go with product updates. It would be nice if the software told you a new version was available and prompted you to do the upgrade to keep everything up to date. At the moment we tend to find out via the support team on the very odd occasion we do have a reason to talk to them. Otherwise, we have found it to work flawlessly and without issue.

I cannot think of any way to improve it. They recently added a web-based interface that runs on a small footprint VM. This has massively improved the visibility of system metrics and made interacting with the product much easier. I honestly cannot think of anything else to be improved, as I rarely have to interact with the Starwind product. My main interaction is with the Microsoft Hyper-V failover cluster running on the Starwind kit and the VMs themselves, which never need any attention from me.

There really isn't much to the software itself. It's not something I use on a daily basis and mostly only when needed. I'd have to say I am usually tinkering with the software maybe on a monthly basis, if that. The only area that the product could improve would be user training. I've had to learn most of the ins and outs by myself or by asking questions when I have to get tech support involved and that's pretty rare.

The overall product documentation and knowledgebase articles could use some rewriting and clarification. Sometimes the documents reference old versions which are unhelpful. Other documents were written and seem to be translated into English so the writing is not clear. I would also appreciate some more technical documents on how the HCA appliances function, how sync occurs, the use of PowerShell on Linux systems, and some other details. None of this would stop me from recommending StarWind though since you rarely need to review documentation.

When reaching out to support by phone they do have a slight accent but are not too difficult to understand. An emphasis on security can be improved. For example, things like default passwords were not changed initially until requested. Veeam support was available, however, we often went to Veeam direct for slightly quicker answers. I can't really think of much else, however, if I had to split hairs I'd say that on reboots sometimes it takes a while to sync with the other node. The maximum was 30 minutes, however, 99% of the time it was three to four minutes.

The only issue we have seen is with the StarWind Server Manager. We have had to continually reboot the server in order to use it. This can be frustrating at times when you want to quickly log in and make changes on the fly. I wish there was a way to make the server more stable so that we don't have to perform a reboot every time. Honestly, we do not have too many complaints. The best part about StarWind is the customer support. The guys on the team are great and have been willing to work with me at any time of the day as well as on the weekends.

The monitoring and reporting console can be improved in the sense that it can have more information on it. Granted, StarWind takes care of the monitoring/maintenance. However, as an end-user, it would be nice if, when you login to their console, you could be able to see information about the underlying hardware and how the cluster is performing. It would be nice if they created hardware integration packages for various hardware OEMs such as Dell, HPE, Lenovo, etc., so you can manage the hardware via the StarWind console and be able to run firmware updates.

CSVs require the storage to be configured through iSCSI, even though the storage is local. It imparts a small performance hit that's easily overshadowed by the performance increase from getting off 1GB ethernet. The Virtual SAN setup also requires more actual storage than a traditional SAN. My two-node cluster means I need double the amount of physical storage before any RAID is taken into account. A two-node cluster with RAID 1 means one-quarter of the storage you purchase is available. Even though the support was great, getting the same email three to four mornings a week with the same issue got a little disheartening.

We struggle a little with capacity management with the StarWind Hyperconverged appliance. An improved front end for managing capacity may be beneficial. I find it difficult to view which hosts are using the resource, and scaling the allocated resource is difficult from the management console. Some integrated tools for managing partitions and disk space would be very beneficial if possible. Integrated reporting tools for capacity and resource management would also be beneficial allowing us to quickly spot over-resourced hosts.

We didn't know that there was a web management interface as a free option. Once I found it on the Starwinds homepage, they helped me to install it. The web-managed interface (a separate VM) could be better. In the version we are now using it looks like there are functions that don't really work. It could be a local problem. We will upgrade both the HA and the managed VM very soon. We would like to see a more complete web admin management tool where you can manage everything from the HCA to the VMs.

Shipping options should be explained in detail and offer more white glove delivery options even if it comes at an additional cost for some SMBs without docks. Our only pain point has been dealing with the freight shipping initially not showing up when promised, then showing up unannounced, and then sending us some small fees afterward due to the fact that we made them wait (since we weren't expecting them). Starwind has taken care of all of the fees after the fact, however, and they were more than accomodating in resolving the situation. We would like more visibility into the reporting that Starwind uses to monitor our solution for errors. We'd like to be able to receive the same data and notifications that their team receives.

I really don't say this very much, however, I actually don't have many issues with this product. However, if I had to think of something, some people might not like the fact that a lot of the engineers tend to have relatively thick eastern European accents, and that might be an issue for some folks. If I had to pick something to add to the product, it would be nice if you could have more than one user account on their command center VM. It's not really a big deal, however, that would be a good addition.

The only thing I would like to see improved is the level of documentation. I had some initial difficulties understanding the solution which could have been solved via additional documentation aimed at newcomers. The solution does admittedly have many moving parts. I've found that it has been helpful to have the support is included and I have been able to get them to help with anything I have needed. That said, sometimes, the accent from the support technician is a little strong and hard to follow.

We would like more complete documentation to manage on your own without support - even though they are great to work with. Sometimes I like to have a better knowledge on my own in case an extreme situation happens that needs to be fixed immediately. The management console could use a facelift. It appears to do everything needed, but sometimes it is not clear what various options are or do which could be tied back to the documentation. A complete single pane of glass for monitoring would be helpful to self-solve issues without involving support and logs.

We have to pay for support, which is high-end support. That can be expensive, at least for us. It may not be that much for others.

I can't think of anything that is missing. Everything has been top-notch from start to finish. Every interaction we have had has been 100% positive. If you are hard of hearing some of their team has strong accents, but I have never had an issue understanding them, nor has any of my team that had a bad interaction with them. Most meetings are done via Zoom, and I'm sure you could switch to a text based conversation if needed as they seem like a very accommodating team. It's not something I would worry about if you are a potential client.

They have a pre-install questionnaire and it asks about network addresses and such. I made the mistake of thinking they were asking about my current setup. They received that information and applied it to the new servers. When they arrived, I noticed it before we patched them in, which would have caused several IP conflicts. I think that the pre-installation questionnaire was a little confusing and ambiguous. I could have asked for clarification but I just filled it out and sent it. It would be nice if they were just very clear on what they were asking for.

Communication could have been improved during the implementation process. I've also been looking forward to further development of their management console which has been slow to materialize. StarWind is promising full cluster management and backup integration in a single-pain-of-glass console. Possible new features could be CSV-level snapshot capability, Veeam integration, and maybe a more straightforward setup. Granted, you don't have to worry about setups with the HCA, but if you want to implement StarWind vSAN in a lab to test it is a tedious setup process.

My only concern was with the terms and conditions of the proactive support. Our legal department didn't accept them, so we weren't able to take advantage of it. Having the heartbeat monitor working would save some time, by getting issues resolved by support as soon as, or possibly even prior to them happening. I would request that the terms and conditions be reviewed and made more acceptable to corporate security so that we could finally turn on the proactive support feature, as it is included with our support agreement.

I think some performance metrics would be nice to see, especially on the storage side. Since we have a decent amount of storage on our cluster, 80TB, knowing what our IOPS and typical usage is would be handy. There are times where something 'feels' slow so it would be nice what was happening on the storage side. I do wish that if we needed more compute capacity, that we could do it without needing to invest in a lot of network equipment. That is hardly StarWind's fault, but just something to think about.

We did have some issues with shipping the product to one our sites. The company shipping the servers lost one of the nodes in transit. But StarWind was quick to order, configure, and ship out a new node so we could meet our install deadline.

Multi cluster support for their Command Console would be very beneficial. Currently, you can only work with a single cluster at a time. The console is new so I expect much growth in this area in the near future. The other area that could be improved is the tech support locale. Currently, all of their tech support is located in eastern Europe so if you have any issues understanding thick accents it may be a little frustrating. With that being said they have never got frustrated with me asking them to repeat themselves, they've always been very patient.

When we purchased the StarWind HyperConverged Appliances, they shipped with the Windows-based vSAN solution. Since then, they have released vSAN for vSphere, which is based on a Linux VM and would save us money as we would be able to get rid of the two Windows Server Licenses. In the future, it would be nice to be able to migrate from the Windows vSAN to the Linux vSAN without having to do a full restore from backups.

One area for improvement of the solution is that I had to get the HCAs with Windows Server installed to install the StarWind SAN software on, which I really didn't want because of the extra maintenance or overhead, as well as viruses, etc. It's going to take time for them to get their Linux implementation to that point. They already have Linux but it's not as mature and they don't really support it on HCAs. They have it for individuals who want to use it on their servers, but not on HCAs. With Windows, there's always that fear that, if you add any software to it, if you need to configure monitoring software or the like, DLL conflicts and blue screens can result. Similarly, if you use Windows Update, you can get blue screens. Or, there have been times where an antivirus company has made a mistake regarding its virus definitions and it took down the server. The antivirus blocked or deleted a legitimate OS file that it thought was a virus. So I don't run antivirus on the Windows Servers VMs that run the StarWind SAN software. At the same time, I've had to configure Windows Firewall to block everything and only allow any kind of traffic going to the server. The only thing I allow is just Remote Desktop so I can manage it. But even Remote Desktop, in the recent months, has had exploits. I keep on having to do Windows Updates. I prefer Linux because it's not as targeted. Don't get me wrong; it is targeted for viruses and all, but not like Windows Server.

The only real flaw that I have seen so far is this hard drive that was accidentally ejected because when it was received and added back into the RAID. There was an error there. It was not added back into the RAID correctly, so I have an outstanding hard disk. Apparently, a guy just knocked it with his hand as he was in my office, so it was just a small eject. He said that he didn't crash into anything. That is the only thing that has reared its head. The support team was straight on it. I have people coming out this week to replace it because remotely they couldn't add it back into the RAID. I think maybe the HD got corrupted. I have all the ports I need in the back. When you're sitting them next to each other for replication between HCAs, it's quick because it has these dedicated iDRAC cables in the back. However, this means I can't have them in separate locations. We could run it through the network to replicate the regular gigabyte Ethernet, but that would be quite slow, especially with the setup. I don't really know how you would change this because I've got a large site. My original on-prem server is quite far away just in case there is a fire (or whatever), so the other one could pick up the redundancy. Having them next to each other defeats the purpose slightly if there was damage localized here, because I would lose both of them at the same time.

That situation, where Dell EMC servers were going down, has been my only real difficulty. I do understand that we were using refurbished Dell EMC hardware, so that may have played into the difficulties we were having. But at the end of the day, it ended up being something that the wider audience of Dell EMC was actually aware of as an issue. Neither the StarWind technicians nor the Dell EMC technicians were able to actually identify that problem sooner than a week or so. I found after, doing my own diagnosis and my own technician work, that there was actually a solution out there that many people Dell EMC's forums were aware of. The communication between Dell EMC support and StarWind support, in that particular scenario, left something to be desired, for me. I did express those concerns to StarWind and they were very responsive to that. They seem to really appreciate the feedback. I'm hoping that there has been a change that has already been enacted by them as a result.

A past problem that they fixed was related to split-brain syndrome. The only thing that is lacking would be a fool-proof GUI for system administrators. At the moment, the initial configuration is very technical and error-prone. That is the reason StarWind does it for you as a service, which is a great thing. But it would be nice if we could change or rearrange storage assignments ourselves.

We were slightly disappointed with the hardware footprint. We were led to believe, and all the pre-sales tech information requirements pointed to the fact, that it was coming on Dell hardware. Then it came on bulk servers. They asked for some email addresses for iDRAC and the like. We thought, "Oh good, it's Dell. We're familiar with that kind of hardware infrastructure." Our other servers here are Dell so we know how the Dell ecosphere works. But then, these weren't Dell. These are Supermicro, which, when you boil it down, are the same Intel parts. But it's a little reminiscent of putting together OEM PCs. That's how the servers look. But they're in and they're working. What you're not paying for, and that may be why it was £36,000 instead of £110,000, are those Dell Concierge services. They've got a well-rounded, iDRAC infrastructure and we could integrate it into our other stuff. We're all used to how all the ILO stuff works on it. But here it's, "Oh, Supermicro. It all looks a bit '2002.'" It's not what we weren't expecting but it works.

I wish I understood what goes into the StarWind software a little bit better. To me, it's kind of magic the way some of it works. As an IT professional, you don't really want things to be magic. I do wish there was a little more "Here's how it works." There could be more documentation given to administrators to know, just in case you have to troubleshoot this by yourself, what you should look out for.

This is just being nit-picky but the only thing I have run into is that I did want to add more hard drives into the host, so that we could look at doing a RAID 10, and the hard drive prices were pretty expensive. I think they're just getting pricing straight from whoever supplies their hardware. They do have the Dell EMC guarantee where, if you have hardware that goes out, you will have a replacement there by the next day. So if one of my drives goes out they'll have a new one to me by the next day. But I compared the price of one hard drive - I can't remember what size it was - and the cost was about $700. I could buy one like that from a Best Buy for $100. Obviously one is enterprise-level and one is just a personal-user-type of drive. They're not apples to apples but the price difference was still pretty significant. I was expecting more like a $300 or $400 difference. Again, that's pretty nit-picky and I don't think it has anything to do with StarWind itself. I think it's more on whomever they work with for their hardware.

This product is not one hundred percent enterprise-ready, so it is more suitable for SMB. The price could always be reduced.

The only thing my team has recommended improving on is possibly a StarWind-customized GUI to monitor the overall system health, similar to 5nine Manager. There is nothing else I would recommend improving because everything from sales, installation to post-install service for the past year has been great.

It's been a few months since the implementation, and so far, the only improvement I'd like to see is the addition of a web console to manage the clusters instead of a client to install.

Better overall monitoring software. Maybe integration with Windows Admin Center was a good direction to go with on monitoring software.

It could potentially be less reliable due to the Hypervisor, and the cluster relies on Microsoft Windows Server. However, we have not had any issue since putting them in production 12 months ago.

None so far. StarWind has met our needs, exceeded our expectation, and increased our system up-time and availability.

We are happy with the product, and my only suggestion would be a better process for an unplanned power outage.

A desired feature or service would be the ability to have a hardware subscription plan that ensures routine hardware updates in conjunction with hyper-converged software.