Critical Software Features Of Storage Software

By Author: Ted Starsky
Total Articles: 11
Comment this article

At first blush the job that storage software performs sounds pretty simple. All it needs to do is move data from a storage device to servers that use that data. But after years developing storage software Open-E has learned just how many jobs storage software needs to perform. Let’s go through some of them to see what it really takes to make a storage area network (SAN) or a Network Attached Storage (NAS) device work in the real world.

First and foremost the data needs to be safe from loss. This can be accomplished with several different methods. The common ones are often used together to provide a high margin of protection. Storage software needs to offer a means to back up the data. This way it is safely copied and can be restored in the event of a hardware malfunction or worse a catastrophe like fire, flood or earthquake. Other ways of making a safe copy include replications, especially off-site replication. This gives you an up-to-date copy of the storage media.

Because storage software manages business-critical ...
... and often sensitive data, it must be safe from unauthorized access. For NAS (network attached storage) the storage server may be visible on the local network and even directly on the internet. Unless robust and proven access control is provided the data could be access by unauthorized employees or even hackers on the internet. Storage software should interface and integrate with existing user and permissions systems such as LDAP. Or in the absence of external user information it should provide a built in system for defining users.

In any computing environment performance is an issue. Storage software needs to provide the level of performance that interactive or compute intensive applications require. This requires network and disk interfaces that deliver the needed bandwidth and the software needs drivers to support these hardware devices. Today this list might include Fibre Channel HBA’s, 10G Ethernet adapters, 6 Gb/s RAID controllers or SSD Cache. Also the software needs to be efficient so it can handle the system IO to keep up with fast interfaces. Storage software needs to be tuned by its developers to run fast with a large number of software techniques for optimizing performance.

There is more than one method for accessing stored data on a network. Users might be accessing the data through the web (http) or FTP. The data might be shared over the NFS protocol or SMB. And in some cases it may be desirable to use iSCSI. Apple based environments may need Apple Share. So you can see that storage software must support many different protocols. Along with each comes configuration and set up options, making the job of the storage software all the more important and critical.

As the storage software is running it must be able to monitor and log load levels for the network, CPU memory and disks. If there is a problem it needs to inform the system administrator so appropriate actions can be taken. In some environments any downtime at all would be a serious problem. In these cases the storage software will need to take action to preserve continuous operation using a fail-over capability. This is a configuration using two storage servers where a secondary server takes over in case there is a problem with the primary server.

So we see that even at a quick glance that storage software has a lot of jobs to accomplish. High quality and mature storage software can make setting up and using NAS or a SAN a great experience. Conversely without many of these features, administration and successful operation of NAS or a SAN can be a big headache.