As we continue our test drive of the Gridstore Scale Out NAS we will test how easy it is to expand the system. After all one of the reasons you get a scale out storage system is because one day you are going to want to grow it. The ease of expansion is a key component in the success in any scale out system.

Before we get into the expansion test, lets provide a quick update on how things are going so far with the Gridstore Scale Out NAS. One of the advantages of our pro-longed testing method is we get to see how a device holds up over the long haul. We have been running the unit in the lab for almost three months now. Other than one planned software update we have had no unexplained outages.

While we will get into our torture test in an upcoming test drive, we have not been kind to the systems thus far. We have pulled power from the unit without proper shutdown several times and have let lab temperatures get to over 90 degrees and the systems have had no issues. In the past we have had other systems in the same circumstances; inexplicably reboot, turn themselves off or just drop their network connection. The Gridstore NAS just keeps running no matter what we do to it.

The system is written to by a variety of clients everyday, it is holding some of our production data and it has several worker applications that continually are writing thousands of files of varying sizes to the NAS. So far the unit has proven to be reliable and easy to use.

Expansion was equally easy. To add additional capacity you add more nodes to the system. To do this you connect additional nodes to your network and power them on. After a few moments the Gridstore Management Console finds them on your network. As you can see in the video below, adding the new nodes to your storage grid could not be more simple.

With the new nodes added in, additional capacity is now available to expand the previously created volumes or to add new volumes. In the video below we perform both tasks. Both processes were easy to do and the Gridstore MMC wizard walked you step by step through them. It does take a while to expand a large volume so this is something that should be done after hours. Gridstore plans to support online volume expansion in the future but for now expansion is relatively easy.

Another important reason to add more nodes, other than just to add capacity, is increased reliability. With more nodes you can take full advantage of Gridstore's RAIDg so that data can be accessed even if "n" number of nodes fail. As we show in the above video this is something that you can set at the point of volume creation or at the point of volume expansion and changing this setting is easy.

We particularly like that the level of redundancy can be adjusted on a per volume basis. This further makes the NASg ideal for mixed workloads. For example backup data can be sent to the system with low redundancy since it is already another copy of data. Then home directories can have a modest level of redundancy since it is not typically business critical and can be re-created. Finally business critical data can be written to a volume with a high level of redundancy. Our example would be the video files that we create. They can't be re-created and so their loss would be very costly so we store them on a volume that can support 3 node failures.

The Gridstore NASg continues to impress. It has been reliable, performs very well and has been easy to use and expand.

Gridstore is a client of Storage Switzerland