What is our primary use case?
Hyperconverged lab and testing environments fulfill several special requirements for use cases. We use it for demonstration environments and for virtualization and containerization. We have a rather smaller footprint, ranging from two to three nodes over the years.
Mainly, we have an edge use case. Another use case is refurbishing and second lifecycle evaluation, specifically I/O and disaster recovery capabilities, which are important in these lab environments.
Modernization could also be counted as a use case since modernizing data centers on small-scale or edge deployments is becoming increasingly important, even for smaller businesses.
How has it helped my organization?
For my organization, StarWind vSAN has been a first of firsts in hyper-converged storage and has been, first and foremost, a technology demonstrator that should disclose and present the advantages of modern data center constellations.
Over the years, it has also functioned as a lab environment and customer demonstrator specifically for refurbishing old hardware and educational objects for our internal apprentices, effectively transforming our mindset towards modern data center infrastructure. It also served as a proof of concept for refurbishing and cost-reducing existing infrastructures.
What is most valuable?
I have found the overall ergonomy and manageability outstanding during normal functioning. Specifically, the console GUI's monitoring and configuring capabilities can be counted towards that.
Even when not administrated via GUI, the comprehensive Powershell automation module remains, which comes with templates prepopulated for the most common automation tasks.
Furthermore, the solution provides great performance for the price it is listed with and can even run properly without proper hardware, however, it is even more efficient and performs better when procuring proper hardware is possible.
What needs improvement?
My impression has been mostly positive. However, there have been some pitfalls. For example, it runs until it does not - and disaster recovery documentation is sparse and mostly unclear. This consequently follows a long learning curve and can take some time until this can be done efficiently.
The feature log-structured file system has been removed. While I have not used it, I know some colleagues who have reported some performance benefits under certain circumstances.
Additionally, vSAN does not offer extended SAN features like inline deduplication, compression, or S3-compatible access, which is part of most modern SAN feature sets, also in hyper-converged scenarios.
For how long have I used the solution?
I have been using several editions of the solution for more than five years.
What do I think about the stability of the solution?
The stability of this solution is vastly dependent on the hardware procured beforehand. Network hardware is especially important, and bad or glitchy hardware (like the Intel X700 or X800 series) can quickly break things. It runs—until it does not. Be sure to learn or consult about the disaster recovery process, especially if you decide on a witness strategy hosted in vSAN.
What do I think about the scalability of the solution?
The software can be pretty scalable; however, after reaching more and more nodes, one may look at StarWind's hyperconverged appliance instead of going through the hassle of administrating separate systems.
How are customer service and support?
They have only friendly and positive experiences, always a helping hand. Also, they are offering proactive support, which, from the experience of some of my colleagues, is an outstanding experience unusual to many professional service providers.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
We did not use a different solution beforehand.
How was the initial setup?
The initial setup was very straightforward and easy, although one could deep dive into the specific protocol configuration options. Also, I would have liked it if some features or procedures would have been better documented at the time.
What about the implementation team?
We implemented it through our in-house R&D team.
What was our ROI?
I will not publish accurate numbers. However, the modernization aspect in terms of lab infrastructure and testing/demonstration environments has an educational benefit, strengthening our consultancy, demonstrating capabilities and our mindset towards data center innovation.
What's my experience with pricing, setup cost, and licensing?
I would highly advise new users to consult the StarWind technical team beforehand.
Also, consider the paid version if you want the ease of management through the console GUI and not just use it for monitoring.
You can also apply for a testing or NFR license if you want to test it. StarWind vSAN is, by list price, vastly more competitive than some comparable solutions. The setup may be straightforward, fast, and easy; however, it can break things down if not done properly. Before setup, be sure to procure proper hardware components that are not known to cause instabilities.
Which other solutions did I evaluate?
Due to budget constraints, Ceph has been evaluated as an alternative. We also took a look at Microsoft Storage Spaces Direct and VMware vSAN.
What other advice do I have?
Consider StarWind's proactive support. It improves the software's already high value and defuses the long learning curve of some processes! When clustering resources, deeply consider your witness strategy! Especially when you can not guarantee 100% stable hardware (e.g., through firmware issues), this can break things very quickly if you have a witness device in the same vSAN.
Which deployment model are you using for this solution?
On-premises
Disclosure: I am a real user, and this review is based on my own experience and opinions.