When it comes to backup and disaster recovery there is certainly no shortage of options to choose from. Twenty years ago, the market for backup software was carved up between Veritas, EMC and IBM. There seems to be hundreds of vendors today?
Start-up investors typically target opportunities with the largest addressable market share. For the last ten years in data protection this has meant the virtual server market, specifically VMware. Standards and easy access APIs have created a plethora of copycat backup vendors all struggling to form a position in the minds of buyers. You would assume this gold rush created value, choice and low prices but in fact confusion, complexity and runaway costs has been the result. Why is this?
Enterprise data protection isn’t easy. To be more precise, getting a consistently good outcome isn’t easy. A good outcome is always being confident of recovering whole systems, not just individual VMs; in a timely manner, with all the data in place. For many customers with mixed operating systems and platforms this requires a lot of planning, daily administration and regular testing. Systems administrators are fully aware of the administrative grind required which makes them susceptible to the headache-busting promises of “snake oil” sales people.
When good enough is not good enough
At Silverstring, we work with clients who are as obsessive about their data as we are and who share our philosophy that good enough data protection is not good enough. We add value by solving the hard stuff, even if it is not the largest addressable market share. Let’s give you an example.
IBM won the battle versus Sun, HP and others in the Risc/Unix wars, but it won a market that has shrunk by a factor of 10x. That said, just like the mainframe, many companies run their most business-critical applications on IBM Power architecture. So, when it comes to good outcomes the time spent on protecting these systems should be disproportionately high. In our experience, investment of money and time follows the high-volume or mass market, rather than the tail-risk of the few, revenue bearing systems which lie outside of the standard deviation “bell curve”. Risk is not evenly distributed, or predictable, and neither should be your investment in data protection.
It’s the recovery stupid
To recover a business process in the event of a disaster requires the coordinated recovery of potentially multiple inter-connected servers. If these servers are not homogeneous this can be a tricky task, requiring significant skill and coordination. Silverstring is working to eliminate the fear and unknown risks that come with recovering complex, mission-critical systems. How are we doing this?
Firstly, we are developing a common orchestration platform to recover multiple, heterogeneous but connected servers, from multiple backup solutions.
Secondly, we have built solutions which leverage the economics of cloud to allow you to self-provision and self-test recovery scenarios from any location, at any time.
Thirdly, since hunting perfection in data protection can be an arduous journey, we offer a fully managed data availability service, backed by our unique, Sleep Easy Guarantee™.
Next step?
Our flagship solution, Alchemis Protect, provides a backup and recovery target platform from which recoveries can occur with minimal manual effort, remotely and on a pay-as-use basis. This makes recovery testing on a frequent basis less time consuming and far more cost effective.
To find out more about our new offering for IBM Power System users, please check out this Silverstring web page – Power DRaaS
