Data as the Gold Treasure of the Information Age
Using OSL Storage Cluster it will be possible to create a online backup of any application, which may be started immediately. Long term backups may be saved to tapes, without having any impact on your productive application.
Data must not be lost. This is one of the central objectives of information technology facilities. But secure information storage is only one aspect of data management. To avoid downtimes of business critical applications, certain data sets must always be available for productive access. Only a operational concept linking various technological and organizational approaches convincingly is able to meet these requirements. The host-based software solution OSL Storage Cluster provides such concepts and the needed tools around it.
The center of a backup concept should be the availability of applications. Running applications require consistent
access to the related data. If hardware fails, this condition is in principle at risk. Redundant systems and
mirror configurations alone do not eliminate this deficiency yet because in case of a disaster running applications
can still not access the required storage. This requires further software intelligence. OSL meets these demands
with an application-oriented solution, where the view of the application is consequently in the foreground.
Productive data of business-critical applications is appropriately stored in high-end RAID systems (primary storage). Using application descriptions and application-aware mirror mechanisms OSL storage cluster creates application images to inexpensive secondary storage (such as SATA RAID). These images are synchronized quickly at regular intervals. If the productive storage fails, the application images are immediately ready for productive access - through continuous mirroring of log files. A recovery in the conventional sense before restart of the applications is no longer necessary. The productive data is simply resynchronized at run time as soon as the primary storage is available again. If a compute node fails, the integrated clustering guarantees that the application can be taken over by another node.
By connecting a backup node, the application images can be easily and thanks to the disk backup concept with high performance (streaming of the tapes) written on tape. The use of inexpensive SATA drives as secondary storage is definitely recommended, as they are working well at sequential traffic loads, which is the regular case for this configuration.