Storing your unstructured data (such as data lakes, data warehouses, and large image/video archives) on the cloud could be the solution to giving users fast, accurate access to the data they need, when they need it. Leveraging the processing power of CSPs may hold the key, according to a recent survey of 300 senior tech executives conducted by software firm Komprise.
Key findings include:
- 87% of IT leaders rate managing unstructured data growth as a top priority, up from 70% in 2021.
- More than half of companies surveyed manage in excess of 5 petabytes of data, up from less than 40% in last year's survey.
- Meanwhile, cloud usage is accelerating, as on-prem-only environments decreased from 20% last year to just 12% today.
And this trend doesn’t appear to be slowing down according to Data Center Knowledge’s recent coverage of unstructured data management.
Navigating Complex Data Migrations to the Cloud
In this year's survey, respondents said that the biggest obstacle to cloud migration was moving data without disrupting users and applications.
"Data center operators have to provide tools for their end users to be able to search data and to build workflow on data," said Krishna Subramanian, co-founder and COO at Komprise.
Self-service tools are particularly important here, Subramanian said.
Users want to not just store data, but they want to be able to use the data, including unstructured and legacy data, she said.
Self-service features for end users and departments was the second-highest priority for unstructured data management in this year's survey, second only to reducing storage costs.
In particular, 65% of respondents said they wanted to use search and analytics tools on their data.
"That is driving the shift to the cloud," she said. "Cloud isn't just a storage location. It's a place where you can run different analytics and machine learning on your data."
Analysts say this type of flexibility gives enterprises important insights when making decisions on the financial benefits of storing object data in the cloud versus on-prem.
How Enterprises Leverage Data Analysis to Manage Unstructured Data in the Cloud
Pharmaceutical giant Pfizer is looking to leverage cloud-based analytics, integrate with other platforms, improve searching, and implement AI and ML capabilities, said Matthew Braunstein, director of hosting data solutions at Pfizer.
"In the research and development side of the house any competitive advantage we can give our scientists has great benefits to the company as a whole," Braunstein said.
As a result, Pfizer has entered onto a cloud journey, he said.
The company has already moved 2.5 petabytes of unstructured data into the cloud, he said, and will be moving more.
And it's not all going to a single platform, he added.
As a result, Pfizer uses data management tools to bridge its off-premises infrastructure in a seamless way. "It's now a requirement in effective data management," he said.
By moving the unstructured data to the cloud, the company is already seeing positive results, he said.
"Our biggest benefits have simply been to effectively use our resources so that we can focus time and money on the things that move the needle the most in bringing breakthroughs that save patients lives," he said.
He recommends that companies start their own unstructured data management process by understanding what data they actually have.
"Having as much information about your data as possible is key to starting an unstructured data strategy," he said. "Without base level analytics it is difficult to understand what opportunities there are with the data and then empower data stewards to be able to take the correct actions."
Pfizer uses data-management-as-a-service vendor Komprise to make its cloud adoption efficient and seamless. Plus, by using data-management-as-a-service, it makes it easier to integrate with high-level functionality, including Komprise's new Deep Analytics Actions, which creates a global index of files and objects across all multiple clouds and systems.
"Our scientific community is looking forward to utilizing some of the newer offerings coming out with Deep Analytics with actions and are working them directly into some of their data workflows," said Braunstein.