A little over two years ago, an article appeared on the Smart Computing website that provided some tips for how to select the appropriate backup software for your PC. Of the tips it suggested, one of the more interesting was its recommendation to select backup software that stored data in a native format. Storing data in its native format eliminates the need to use backup software to recover it since any computer can access and recover the data. But this article was written for PCs. So the question that companies now need to ask is, “What do they need to consider before selecting a product that will allow them to store replicated data of its enterprise production servers in its native format?”
Before answering that question, it’s first important to understand why storing backup data in its native format works fairly well for PCs and laptops. First, the amount of data that actually changes on a day-to-day basis on most PCs and laptops is minimal so the performance overhead associated with replicating data on PCs and laptops is also minimal. Second, the replicated data only goes to one target device (network drive, secondary disk drive or even a USB flash drive) so it is fairly easy to configure. Finally, the individual responsible for managing the software is often the same person using the software so there are minimal or no concerns about central administration.
Now let’s extend the discussion surrounding storing replicated or backup data in its native format to include production application servers. Can or should organizations look to replicate production data from these servers from production disks to backup disks and store the data in a native format? The answer is yes but organizations need to carefully consider what tool they use to do this since how it is implemented and managed differs from how it is done for PCs and laptops.
The biggest difference between the two is that the software that enterprise application servers require must be much more sophisticated. For instance, enterprise replication software needs to support many more operating systems such as Linux, Windows, and VMware, it needs to integrate with critical applications such as MS Exchange, MS SharePoint and MS SQL Server to create multiple, consistent recovery points and it may need to replicate the data to multiple disk targets.
To accomplish this at an enterprise level, enterprise replication software needs to scale so it can concurrently and centrally monitor and manage the replication and recovery of this data across multiple physical and virtual application servers. This is no trivial task since it requires the replication software to constantly adjust to changing conditions such as server overhead, network traffic and even outages in the network or on the source or target. Further, the job of copying or replicating the changes to the production data should not cause undue overhead on the server or negatively impact the performance of the application.
InMage Systems Scout distinguishes itself from other products in that it stores replicated data in a native format so any computer – physical or virtual – can use the replicated data for recovery. But equally or even more importantly, Scout includes the underlying architecture that enterprises need so they can scale it to manage the replication across tens or even hundreds of servers while still maintaining a central console to monitor and manage the replication and recovery of the applications.
Storing backup and/or replicated data stored in its native format is not a new concept as it has been used successfully on PCs and laptops for years and is now moving upwards into the enterprise. But before organizations take this jump and implement it on enterprise application servers, they need to look beyond the obvious benefits that storing data in its native format provides and consider how they will configure, manage and scale the software that does the replication so it works consistently across all of their enterprise servers. InMage Systems’ Scout is one such software product that is architected so that it can handle and adjust to the requirements that enterprise applications have while still scaling to centrally manage the movement and recovery of replicated data in its native format anywhere in the enterprise.