How to Create a Data Storage Strategy

Businesses today have more data than ever. In fact, according to IDC, global data volumes are set to grow at a rate of more than 60% a year through to 2025, by which there’ll be some 175 zettabytes of information stored around the world.

Therefore, it's vital to have a clear strategy in place for how it stores and manages these assets, which ensures users are able to quickly access data when needed, while also keeping the costs of long-term storage under control and ensuring the security of confidential information.

There's no one-size-fits-all solution for this. Each business will have its own unique data assets, which could range from large to small video files, but numerous, sensor recordings and financial updates. Therefore, it's important to take the time to plan out a storage strategy and build a bespoke solution that's tailored to your unique needs.

But what factors do you need to consider in order to make these efforts a success? Here are a few key things to bear in mind.

1. Understand your data needs

The first thing to do is to audit your environment to determine exactly what type of data you have and what the requirements for each type will be. Not all data is created equal, so you should understand the business value of each piece and what will be needed to make the most of it.

Key questions to ask include:

• How quickly will I need to access this data?
• What would happen if this data were lost?
• How long should it be retained?
• What regulatory requirements will be relevant?

Answering these questions at the start of the process will help you identify your most valuable data, the items which need especially high levels of protection and oversight, and which can be rated as a lower priority.

2. Have a plan to deal with unstructured data

A big challenge for any business today is dealing with the growing amount of unstructured data, which is now estimated to account for as much as 80% of all business data. This includes video, email contents, voice recordings, web pages, presentations  - essentially anything that doesn't fit easily into a traditional relational database.

However, this data can be used to gain significant insights if it’s analyzed properly, so the storage platforms you use must offer a high level of performance in order to allow big data analytics tools to access it easily.

3. Don't overlook compliance

With the introduction of new regulations such as GDPR and the California Consumer Privacy Act (CCPA) - not to mention the range of industry-specific requirements - all businesses need to make compliance with privacy and data protection laws a key part of their storage strategy.

As well as ensuring there are tough encryption and access controls to prevent unauthorized viewing or copying of data, you need to ask if data can be easily accessed in response to freedom of information requests, and if you’re holding any unnecessary data. GDPR, for example, requires firms to minimize the amount of personal data they hold and ensure it’s not kept for longer than necessary.

4. Take a tiered approach

An effective data storage system should include a variety of solutions depending on how valuable the data is and how often it’ll need to be accessed. Getting the right tier for each type of data is vital in both ensuring the best performance for your system overall, but also ensuring your data storage strategy is as cost-effective as possible.

In general, Tier 1 should encompass mission-critical applications, being reserved for data that demands fast read and writes and where latency may be an issue. Technologies like in-memory and solid-state drives are most useful here. Tier 2, meanwhile, typically uses disk arrays and is for more static details such as 'cold' data and historical financial information, while Tier 3 is for archives, and can use slower tape drives.

5. Focus on the long-term

Building an effective storage system from scratch can be a daunting process, and one key concern will be the upfront costs involved, especially for higher-performance solutions needed for Tier 1 data. However, focusing too closely on this capital expenditure will be a mistake, as it can mean you end up overlooking ongoing expenses that could significantly add to the total cost of ownership.

For example, if you're looking at cloud options for long-term storage and archiving, you need to factor in the ongoing costs of your subscription, and it can be tricky to predict what this could look like ten years down the line. On-premise tape drives, on the other hand, may be a larger initial investment, but will have much lower ongoing costs.

There are a wide range of ongoing factors to consider. What will the long-term power costs be? How easy will it be to scale up your system as data volumes grow? What’s the expected lifespan of your hardware? Even the cost of renting or building floor space for your storage system will need to be taken into account if you are to develop a solution that stands the test of time and ensures the costs don't spiral out of control.