Making The Unpredictable Predictable: The Challenge Of Data Restoration
While technology news outlets tell stories of moving to the cloud, big data and cybersecurity threats, IT teams deal with the reality of an information-driven business world: the challenge of storing an ever-increasing amount of data. We have an insatiable desire to keep everything from email to spreadsheets, just in case we might need them in the future. While tape and hard drive storage costs continue to drop, the growth in data is far outstripping that rate.
As if dealing with the technical challenge of managing all of this data were not enough, IT is faced with another challenge: how to handle data restoration for legal and regulatory events. These kinds of events are unpredictable from both a timing and scope perspective; wreaking havoc on IT budgets and planning.
Financial Impacts of Data Restoration
Unfortunately, unexpected data events are not discretionary, and IT has to find a way to handle them regardless of competing priorities. When the data is readily identifiable and available, the disruption may be minimal; however, sometimes the location of the data may be unclear or may be on obsolete formats or media. In these cases, IT can face a very difficult challenge. For example, where is the data located (in-house or in the cloud), is the data available in a human-readable format and how quickly can it be found and restored into a format presentable in court?
Typical technology lifecycles are three to five years; typical legal retention requirements are seven to ten years.
A fundamental issue is the disconnect between the speed of technology change and the legal need to retain data for many years. This means IT may have changed technology twice during the retention lifespan of a typical piece of data. When technology vendors change data formats, IT may be left out in the cold with the need to retain old hardware or software to access data stored on older formats.
Since IT is required to capitalize most enterprise hardware and software, retaining multiple generations of technology can create a depreciation burden on already stressed budgets. Even if the fixed cost of depreciation is not an issue, the operating expenses for maintenance may be excessive. In either case, precious dollars are diverted from meeting the demand for new business solutions.
IT is Designed for Data Restoration
IT organizations are designed with two basic teams: infrastructure and applications. Infrastructure teams may be organized by platform (desktop, computer, storage, networks, etc.) or by service (email, unified communications, file and print, backup, security, etc.). Application teams are typically organized around a business function such as finance, human resources, procurement, etc. The infrastructure team is usually responsible for handling data restoration events.
Just as other parts of the business have shed non-core services to third parties, IT is going through this same transformation.
Whether organized around platform or service, the team has a horizontal view of data; i.e., each platform or service may have its own backup and archiving approach and technology. Data restoration events tend to be vertical; i.e., data about a person or subject that may cross every platform or service. Therefore, IT is at an immediate disadvantage when faced with data restoration. Compound this with having to go back to data on obsolete media, and you can see why IT has to jump through hoops to handle data restoration events.
The goal is to make the unpredictable, predictable, but to do so, IT may have to step back and take a hard look at whether data restoration should be handled in-house.
Data Restoration Takes Special Expertise
With the exception of email, it is not easy to identify data that may be relevant to a data restoration event. For example, a Microsoft Word file will have a creation and “last updated” date, but any other identifying information is unreliable. This is because people often create documents from templates and never bother to change the metadata describing the author or description. Even the “last updated” date is unreliable, as it changes if the file is copied from one place to another.
There are specialized forensic tools that can be used to retrieve data from damaged or obsolete media. These tools are not found in most IT teams, nor is the expertise needed to use them effectively. It makes more sense to work with a trusted partner for these needs, rather than trying to maintain the expertise in-house.
Moving Toward Service-Based IT
With the rise of cloud computing, Software-as-a-Service (SaaS) based applications and managed service providers, today’s CIO is becoming more of a “service broker” rather than the direct owner of technology. While outsourcing IT used to be an all or nothing proposition, the CIO can now be selective about what stays in-house and what can be handed off to service providers.
Data restoration is a prime candidate for a trusted partner, especially if that partner is already helping to manage archived data for your organization. While in some cases it makes sense to hire data restoration services on demand, the reality is that the need for this kind of service is only going to rise as more and more information becomes electronic rather than paper-based. Having a service provider that understands your environment, understands the nuances of legal and regulatory requirements and can respond at any time will be a key part of your IT’s service portfolio.