I use the tool by myself. I do recommend it to other customers. I am a Microsoft MVP, operating most likely in the functional areas or just the KPIs. I do recommend the tool to others as it is a good product.
The features of Azure Stack HCI that have proven to be the most beneficial for optimizing our company's data center operations revolve around its performance. Recently, I tested a four-node solution with NVMe, which is the sixteen generations tool that falls under the 16th generation solution from Dell, and I was able to do five million apps, so it offers pretty good performance for me.
Speaking about the tool's challenges, I would say that if you know the product, you probably know why there are challenges in it. The challenge now is that Microsoft has too many things to do with the management of Microsoft Azure Portal. There are just too many things you can do from Microsoft Azure Portal. So you have to do a lot of things in a traditional way with PowerShell, which works fine for a lot of customers, but for a lot of customers, it is not enough. The biggest challenge with the tool is not being able to manage everything from Microsoft Azure Portal.
In the future, I want to see the tool offer better integration and multi-cluster management in Microsoft Azure Portal. With multi-cluster management, you can view all of your virtual machines across all of the clusters you have and maybe move the machines between the clusters, which is something that I would love to see.
I have been using Azure Stack HCI for nine and a half years.
I have never used the solution's technical support as I usually have my own path, but I have heard that they are good. The support is cheap. You can pay 100 bucks and have support from Azure. You just create a ticket from Microsoft Azure Portal, and it is easy to get support.
I have specialized in HCI tools for Dell for three years now, so I know a little bit about Azure Stack HCI currently.
When I talk to the customers, I see that cost savings come from the licensing part because a lot of customers already have Microsoft Software Assurance, which one can get with Azure Stack HCI. If you are interested in using the Azure Arc, it is free for Azure Stack HCI. If you are in a VMware world, then there are a lot of licenses for VMware tools, so I guess Microsoft is comparatively cheaper.
Speaking about how Azure Stack HCI's scaling capabilities impacted our IT infrastructure management, I would say that the tool's scalability part is not that good because you, on paper, you can scale up to sixteen node clusters, but the optimal size of the cluster is somewhere between five to eight nodes or even five, six nodes at times. If you have four nodes with five million apps, then it won't mean that you can have eight nodes with eight million apps. You will have four nodes with five million apps and eight nodes with six million or seven million apps. The scaling-up process is not linear, and it is not the end of the challenge. I have never seen a customer doing many scaling-ups with the tool. It is quite a robust solution, but nowadays, we are not talking about the scalability features with the customers, as we are currently talking about the benefit of having a hybrid solution because it is connected to Azure, which can then allow you to consume some of Azure's services. It may not allow a user to consume everything under Azure's services, but a few of them can be used.
Speaking about the integration with Azure services enhancing Azure Stack HCI's usage, I would say it is like a promise. When you keep asking what is available from Microsoft Azure Portal, you will find out that it is just the virtual machine deployment, and then you'll find out that the resource bridge can't support the virtual machines. It was more of a promise of what could be done in the future if Microsoft implemented it correctly. Everyone is eagerly waiting for 2024 Microsoft's release, which will most likely be in October.
Azure Stack HCI support for initiatives in AI and machine learning is not something that Microsoft is discussing clearly. You can obviously use the GPUs that you can pass through into the containers, but there are no SKUs in Microsoft Azure Portal that can use the GPUs right away, so you have to attach them, making it kind of complicated, and there isn't a native AI available right now. It is a little challenging because Microsoft has to come up with something, but it hasn't come up with anything yet.
I won't be able to recommend any other product in place of Azure Stack HCI.
I recommend the tool to others.
If you are into open-source tools, you would probably go with KVM, and if you know nothing about Microsoft, then Azure Stack HCI would not be the right solution for you. For most companies or corporations, Azure Stack HCI is suitable for them. It offers nice performance and specifications and is relatively cheaper than other tools.
I used to work at Microsoft three years ago as a PFE. I used to write for workshops on how to use or what technology is used in Azure Stack HCI. I used to run the workshops. I trained internal people on the tool. I know a lot of big brands because I worked with the product managers at Microsoft, and we used to shape the products together. Even now, I still work closely with them, so I know a lot about the tool.
I rate the tool an eight out of ten.