High Availability–What It Means in a Managed File Transfer Environment

Don Miller

High availability can invoke feelings of insecurity when it comes to evaluating your managed file transfer software resources, but it need not be such a scary prospect.  In its simplest form, high availability allows you to ensure that your system will be operational as close to 100% of the time it needs to be.

This can be accomplished in one of two ways.  You can have the software installed in a clustered environment which would balance the load across multiple servers or images, or you can have a redundant system that lies dormant until the need arises for it to take over and execute when the primary environment stops.

There are two basic cluster types, Active-Active and Active-Passive.  In an Active-Active cluster, two or more instances run concurrently sharing the load and accessing shared database and shared file systems.  This type of setup has several good features, the first being load sharing.  If you have considerable traffic coming in and going out, having two or more servers increases your throughput, thereby allowing you to process more transactions while using your CPU and Network bandwidth to the best of its abilities.  This method is also the most expensive, however, since it usually requires a license for each server in production mode, similar hardware, shared database and file systems.  But if your volume requires it, it is the best way to go.

The second cluster type is an Active-Passive environment.  With this type of installation, one server is passive–i.e., not running but waiting for an event to occur before becoming active and taking over from the primary server.  The advantages to this type of installation is that the passive server can be used for other applications, and licensing requirements can allow this sort of configuration at a reduced cost since it is only used when the primary system is out of service.  There are a couple of different ways this type of installation can be configured.  One choice is a hot spare, where the system connects to the database and file systems, and only needs to be triggered when the primary system fails.  This provides the least amount of downtime and can rapidly take over should the need arise.  The second method is a cold spare that would require manual intervention should the primary system go down.  This method could use synced copies of the database and file systems and would probably cause the loss of data should an event occur while the systems are not in sync.

To state the obvious, if your business requires that your system be available 24 hours a day, 7 days a week, the best choice would be a clustered system with a hot spare located separately from your primary system.  With this configuration, you should have an unobstructed flow of communications traffic.  As long as your remote site is synched to your primary site, the likelihood of lost data is drastically reduced and your systems should be available within seconds of a primary system failure.

Which is the best?  Only you and your managed file transfer software vendor know for sure.  Determining the length of time your system can be offline is for you to determine.  You know your business and how critical is the transmission of data.  If you normally transmit data during the business day, and very rarely have transmissions in the off hours, an Active-Active system would be overkill, and an Active-Passive system is best.  This allows you the ability to take your primary system offline for maintenance should the need arise and have your passive system take the load for a short period of time.

On the other end of the spectrum is a business that moves data at all hours of the day or night.  In that environment, having your system offline even for a few minutes can result in severe backlogs and or financial penalties.  So you would probably want an Active-Active cluster with a duplicate cluster located in a remote data center and configured to take over immediately in the event of a primary system failure.

So there you have it; the different types of high availability systems and some points to consider about each.  bTrade’s secureXchange MFT suite has each of the above capabilities included, along with a relay component to install in your DMZ if you wish to secure your installation.  All that, combined with our proprietary encryption algorithm known as TD Compress and all available encryption nodes, will provide you with the knowledge that your data is secure.

If you have questions, please contact one of our data security experts by sending a confidential email to info@bTrade.com