We’d like to think that we can rely on Cloud servers and systems to take care of our data and confidential information well, with the use of the redundancy principle. Redundancy is adding an extra element into a system to ensure continuity of function in scenarios of an element failure. But is the answer to the reliability question that the system will run smoothly even when one element fails if a backup will automatically pick up the pace while the failed system is in recovery? Reliability is a complicated issue that needs to be studied significantly before giving it a simple solution of redundancy. By making systems from the primary to the secondary storage systems redundant does not answer the reliability question but rather brings in more risk from errors, not to mention is costlier.
Before Cloud, businesses and other organizations use hardware – CD, portable hard drives and other physical storage – this was the fail-safe measure from loss of data in cases of catastrophic events including flood, fire, other natural disasters and manmade disasters such as break-ins, destroying the initial system. Cloud storage services made this easy for you, by storing your data in a remote location. They designed their data storage facilities in such a way that it prevented damages from catastrophic events. In this measure, your data is safe and decreases the risk of loss of data.
Data Corruption through Cybercrime and Tech Failures
In the case of data corruption through hacking or other cybercrime and technical failures, redundancy proves to be less effective as a risk management measure. First, it must be understood that there’s such a thing as Retention Policy. This is an organizational method of retaining data – including the number of variations or variants of the data will be kept, duration or period it’s kept and the final fate of the data at the end of the period. Retention policy is applied both within the enterprise or organization and on Cloud storage providers. Although a redundant program allows access to corrupted data, through a backup, the problem lies on whether the data stored in the off-site/redundant program is the same version or variant of the corrupted data.
Cloud storage programs usually keep a fixed number of versions – 3 to 30 variants on a fixed duration or period of 30 to 60 days. If your system is corrupted within 15 days of the 30-day period, you are allowed to access the backup and fix the corrupted data within 15 days and retrieve the version you need. If corruption is not detected or fixed correctly within this period, and the period of data retention is only 30 days, there is the risk that all will be lost – even the uncorrupted data in the backup. This is where redundancy within a Cloud system becomes risky.
Reliability is a complex issue in an environment where it is necessary but not easy to achieve because of the different elements present in an enterprise. Being fail-safe is to look into your system, design protocols and policies that will ensure the running of the business ops even when a disruption happens. Redundancy is a tool but not the cure-for-all. Consider redundancy a troubleshoot while you are figuring out, clearing out, and fixing the mess, rather than the fix itself. Reliability must consider all aspects of probable failures even with a Cloud storage system. Measure and figure out fail-safe plans within each issue, and how your operations can keep running while fixing your issues.