What is most valuable?
What we like is the performance of the equipment. It's really much better than hybrid aggregate or machines with flash cache. We have been using the FAS series for a long time and it's still performing well. First we started with 7-Mode. Then we moved our databases to clustered data ONTAP. Today we have more than 24 nodes; we have a lot of machines working in cluster mode with all activities on site. It works perfectly.
How has it helped my organization?
We use an ONTAP cluster for the core Oracle DBs. The benefits are performance and the features we use, such as FlexClone to clone and restore the DBs everyday and to check if backups are properly done. These are great benefits.
What needs improvement?
In future releases, I would like to see improvements in performance. That's something that we always need. But the performance that we have right now is really good.
Also, I'd like more features related to All Flash FAS with OpenStack or All Flash FAS with Manila.
What do I think about the stability of the solution?
In terms of stability, it is much better than 7-Mode. In terms of stability and performance, it is a very good machine with very good improvements. The cache layers are warm and everything is solid state; the kickbacks are really fast; better than other solutions. It is really good.
What do I think about the scalability of the solution?
We scale one or two controllers every year by adding an extra part to each cluster if we need it. Last year, we just bought a shelf, but in the previous years, we were increasing by one or two HA pairs per cluster; that's a lot for us. But, it's easy to scale. The most interesting thing we did is we moved volumes across the cluster without downtime and with a minimal performance impact. That's something that we couldn't do in the past with 7-Mode. So that's really good for the company. For a commerce company like ours, we can't support these functions with downtime; it must be while online.
How is customer service and technical support?
Technical support is good, but it depends on the tool that you're using. In the past, we had troubles with DFM and we eliminated DFM from our infrastructure. Support for OpenStack, Cluster-Mode, and 7-Mode is really good. Because they have been doing it for many years. But in general, support from NetApp is really good.
Which solutions did we use previously?
We decided to switch to all flash because we needed better performance and lower latencies that are stable with higher IO. That's something that traditional arrays can’t do.
How was the initial setup?
We normally set up our clusters ourselves. We request professional services from NetApp when we want to add new machines into the cluster. But for other tasks, such as to configure, generate reports, create the aggregates and move databases across the cluster, we don't need technical support. It’s relatively simple.
Which other solutions did I evaluate?
We tried SolidFire also and we liked it. But we don’t use it for Oracle, we use it for OpenStack. We also looked at other companies. For example, EMC, which is a good solution, but it's really expensive. If you compare it with NetApp, the performance is the same. When using NFS, the best is NetApp. For Oracle, we are using NFS. NetApp does not have a competitor for NFS.
Disclosure: I am a real user, and this review is based on my own experience and opinions.
Jan 04 2017