Can Commvault Activate Make Backup Data the Key to AI and ML Success? Posted on : Dec 28 - 2018

One of the most important things a business can do when its using AI products or creating custom AI projects is to make sure all the relevant data can be used as inputs. AI and machine learning are hungry for data and work best when the algorithms are using large datasets. The predictions generally become stronger the more data with important signals there is to work with.

But wrangling and assembling all the data is difficult: in most projects, whether AI or analytics-focused, it’s a huge undertaking to assemble all the data needed. This has never been truer than now, in the era of big data, when companies have so many data sources and formats to work with.

Commvault, a data backup, recovery, and management company, has long understood that for AI and ML applications to work as robustly as possible, companies must have the ability to understand what data they have at their disposal as quickly and easily as possible. Finding and accessing data is the key to creating data pipelines that allow the right data to be extracted for use.

Recently, Commvault launched a new product called Activate that could make finding the best data for an AI or ML application much easier. Activate makes a company’s backed-up data available in a similar fashion to a data lake repository. By taking advantage of the 4D Index that is created by the Commvault Complete Backup & Recovery platform, which captures a complete set of metadata, the backed-up data can be searched easily. Activate also has the ability to create data pipelines that can access the backed-up data to extract relevant information.

I recently had the chance to speak with Commvault's Patrick McGrath about Activate and what it portends for AI and ML success. Based on that conversation, I think it is possible that if Commvault succeeds with Activate, backups could become more than just something used for data recovery and protection, and instead, become an integral part of how companies make full use of their data to achieve the AI and ML goals they’ve been struggling to accomplish with data lakes.

Indexing It All

Activate is able to function as a data repository because it relies on Commvault’s 4D Index of data which powers the entire Commvault Complete platform. This index is an abstract look at all of a company’s data, not just the form of the data or how it is stored, but the nature of the data itself, independent of its storage and format. “With Activate, we repackaged much of the Commvault platform to make it a lot easier to buy and understand. The foundational component is the Activate 4D Index, which is abstracted from the Commvault data platform. That allows us to gain an understanding of the information held within the underlying data BLOBs, sourcing data from backups and archives that may already be under Commvault management, but also directing that content indexing to focus on live data sources that we currently don’t manage,” McGrath said. View More