One of the most important words in our industry has suddenly become “immutability”.
Why? The prevalence of attacks has been pushing this characteristic to the forefront of our minds of course, but hasn’t protecting our data always been important? Why think about immutability now? What’s changed?
A key difference in today’s business model compared to several years ago is the requirement of near instant recovery of operations. In the past, if applications came back online within a day, with 24-hour old data, it was acceptable. Those days have long since disappeared.
There are varying designs and architectures that help implement fast recovery of data, for example:
- Application-level replication or clustering between sites
- Full Active/Active infrastructure and systems
- Replication of primary data between sites
- Replication of backup data between sites
The question to ask is are the separate sites and infrastructures isolated from each other in a way that would stop any type of attack on data? If the redundancies at an application-level fail to provide that type of protection, then we have to think about recovery of data. For business reasons we must accomplish this in the fastest method available, which typically means minutes.
Quick restoration of systems requires that backup data be kept local to the primary data so that it is available quickly. The danger then, is that it makes the backup data vulnerable to attack, which is where the conversation of immutable storage of data comes up. This isn’t as straight-forward as you might think however, because so many vendors are using the term loosely and with products that might not give you what you are really needing. For some, it has become one of those trendy IT expressions that we have seen used in marketing of products ever since computers have been around.
We need to think of immutability of data as a continuation of the discipline of defense layering that is the bedrock of all data security. It is a goal that is to be constantly strived for and thought through by asking the question “what if” over and over.
Fortunately, there are some data storage products are getting very good at keeping data safe in the same location as the primary data and should be used as a way to recover data quickly. The ability to recover data within minutes, and with a very short RPO, is critical in the battle against ransomware and is a necessary be part of your strategy. Our approach at Dataedge is to evaluate these products from many different vendors and determine not only if they work as advertised but also which ones will be the best fit for your specific needs.
One other thing to consider – we see a somewhat dangerous trend in that business managers can hear “we have an immutable product that protects our data”, and they think:
- ok – now we are safe
- we no longer have to worry about that
- check that box off”
It’s human nature to want to think that way, but we must keep in mind that data protection and security is an endless process. We can’t forget about the layers of security that should be used in conjunction with immutable data storage. Why? Because what is considered safe today may have some type of back door that will be exposed in the future, and we must stay a step ahead of the bad guys.
I’ll have quite a bit more to discuss on immutability as it relates to backup data but will save them for future blogs. Check back soon.
#dataprotection #security #securedata #storage #datastorage #cybersecurity #backups #restore
