What is our primary use case?
The primary use case of this solution is for virtualization, data restore for VMware, NFS exports for a backup share, and in the future, we are thinking about using this to update based storage for S3 buckets.
At the moment, we mainly use it for NFS topics.
What is most valuable?
GPFS monitoring is the best feature. The performance is very good and so is the scalability.
The NFS CL solution is impressive because if you have a power loss for any reason from the consumer host, the system switches seamlessly to the next available node in this cluster. This happens without any interruption in the data stream, reading or writing.
When the system is coming back from the maintenance or the reboot cycle, the system switches automatically back to the initial host. This is very good and very friendly. I have never seen this before when comparing to other solutions such as NetApp, and Dell EMC Unity. NetApp is also very good for NFS solutions.
What needs improvement?
The biggest problem is that it is not able to provide block storage. It can consume block storage, you can put it on from the storage and present the span over the fiber channel or iSCSI. The problem is that they have not been able to provide block storage consumable to the customer host. This is only a side-based solution. We can install the GPFS into the customer environment but I don't think that is the best solution.
The initial setup has room for improvement, it is not straightforward.
Spectrum scale is very good from the performance perspective, but from the application perspective, you have to take care if you want to use a database as a service that you can consume. For example, a database, or being able to read and write it, doesn't have a deployment solution. It is missing.
For how long have I used the solution?
I have been using this solution for two years.
We started with version 502 and now 503.1.3.2 the lastest and the greatest version.
What do I think about the stability of the solution?
This solution is stable, it's very good.
What do I think about the scalability of the solution?
This solution is scalable it's very good. With this solution running, you have solved some cosmetic issues. The system is very stable.
The biggest issue we have faced is the complete interruption of the network. We have tested a complete site loss with the initial loss of the complete data center. We tested this and the system was always up. This is very good.
We have 10,000 clients, so there may be 100,000 to 200,000 users. I am not sure.
How are customer service and technical support?
The technical support is good.
At this moment we have a problem with technical support but it is not related to Spectrum Scale. It's with another product called Tivoli Suite, Spectrum Control Site.
Which solution did I use previously and why did I switch?
Currently, we are also using Nutanix, Vison, ScaleIO, and Huawei as an STS solution. The problem is that we want to use one solution in the future and not five.
I have to compare all of these and decide which one is the best for the future.
How was the initial setup?
The initial setup was complex. We had to prepare a lot of the requirements to install the Spectrum Scale. It is not straightforward. We had to ensure that the foundation for the install and that the installation process was proper.
On average for large clusters, the deployment can take one week.
What about the implementation team?
We did not use the help of an integrator, vendor, or reseller. I implemented this solution.
What's my experience with pricing, setup cost, and licensing?
The licensing is based on the number of terabytes.
We have an enterprise license agreement with IBM and it doesn't reflect the moment price list.
I think that it's a good price but it is quite expensive.
What other advice do I have?
We are using this solution in our Opcos and the challenge in my role as an architect is staying informed of what solutions are to be used in the future. What is the best solution to carry that will cover all of the requests that we have and for the multiple cloud solutions regarding STS soft topic?
I am currently making comparisons for this and I need support for this as a strategy for the next five years. I am the lead architect for this solution in our company.
My advice would be to use a consultant or a technical person to implement this solution. It's a better way.
I would rate this solution an eight out of ten.
Which deployment model are you using for this solution?