Is data gravity weighing you down?

Data Gravity?

In 2020, every human generated at least 1.7Mb of data per second, 2.5 quintillion data bytes daily in total. Sources suggest that by the end of 2025 this figure will be as high as 463 exabytes.

Data gravity is a buzzword that was first coined back in 2010 in a blog article written by Dave McCory, an engineering VP at GE Digital. The term refers to the relationship between data and applications, as data grows, they become attracted to each other, like how the Law of Gravity works.

Datasets are getting larger and larger, which provides a problem when trying to move them around, giving them more ‘gravity’. This means the data remains in one place, while applications and other data becomes attracted to it, causing it to gain in mass.

What are the 2 main challenges of data gravity?

Businesses grow through acquisitions and geographic expansion. Even small businesses need to reach new audiences by creating new sales channels, products, and services, and maintaining data wherever a business does business is imperative to the success of the organisation. These businesses must think about ‘data gravity', as it is an important choice when considering where to host and connect data because of these challenges. Connecting data is needed to execute business processes successfully, as well as bringing the user, applications, and the network closer together.

Get articles from our team of experts in your inbox Subscribe

There are two main challenges to solving the gravitational pull of large amounts of data:

Latency - A large dataset by nature requires that the applications that use it, be close by, in its orbit, or it will suffer from latency. Because when applications are close, the better the workload performs.

Speed is crucial to all business processes and increasing the latency as data expands is not possible. Businesses will need to ensure that the workload balance expands simultaneously with the data’s gravity. To solve these applications, need to be moved alongside the data to prevent latency.

Non-portability - we already know that data gravity increases with the size of the dataset, and when the dataset gets larger, the more difficult the dataset will be to move, and when moving large amounts of data, the process is slow and uses multiple resources.

When data is been migrated, data gravity needs to be considered, and businesses need migration plans that include future dataset sizes, and how many services and applications should be determined.

Data gravity needs to be handled with care so that it doesn’t become an insurmountable problem, keeping business operations running smoothly and efficiently. This is where a good data strategy comes into play, including a good data management plan and data governance, allowing data to flow, and mitigating the negative impacts that data gravity can cause.