Enterprise Storage Forum content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
Data classification is an essential aspect of the storage of enterprise data.
In addition to helping organizations manage and search their data repositories, data classification helps them comply with relevant industry-specific regulatory mandates, such as SOX, HIPAA, PCI-DSS, and GDPR.
Data classification has become even more important as the amount of data mushrooms. Unstructured data especially is being stored in much greater volume. Companies now want to store it longer and get insight into that data.
“Most companies don’t even know how much unstructured data they have or what’s in the data they are backing up, taking snapshots of, and making multiple copies of, while absorbing more and more costs related to the management and cloud storage of this unknown data,” said Peter Worsnop, VP, sales Aparavi.
“That creates a recipe for disaster that must and will be addressed in 2023, as cost reduction and intelligent management will come into play as companies adjust to the new economic realities happening globally,” Worsnop said.
Here are some of the top trends in data classification:
1. Data classification grows in importance
The world has never generated more data. In fact, that number is estimated to be about 2.5 quintillion bytes a day.
That staggering number makes data management intensely complicated and costly. Because of this, companies must seek out ways to classify their data more effectively and efficiently. Those MP3s employee downloads do not require the same level of data protection as personally identifying customer information.
More data, however, means more costs and increased security risks on that data from bad actors. Because of this, we’re starting to see data classification become increasingly important to both manage costs and manage risk, according to Mignona Coté, CISO, NetApp.
“Organizations need to know where their most important data is located across their data estate and make sure it has the most robust protection necessary to protect it from all threats – internal and external,” Coté said.
“Once classified, the highest level of protection can be applied to a company’s most critical data, including monitoring for data exfiltration, tier-one backups for rapid restore, advanced auditing/logging, and validation of appropriate access permissions across the board.”
This approach to building protections around classifying data allows organizations to shift from protecting devices to protecting the data itself. Knowing how data should be accessed, by whom, and for what purposes allows for a robust zero-trust model and simplifies security alerting and response.
To achieve this, companies are increasingly reliant on AI-driven natural language processing (NLP), as it delivers contextual data analysis and categorization for actionable insights into their data to address compliance requirements, detect security vulnerabilities, optimize costs, and accelerate migration.
“Effective data classification and tagging is paramount for building world class data protection and resiliency programs,” Coté said.
2. Increasing security concerns and ESG
Travis Johnston, director of market strategy at Folio Photonics, named increasing security concerns and environmental, social, and governance (ESG) as one of the primary data classification drivers. This is due to several factors:
- Ransomware continues to plague businesses and governmental agencies around the world
- A bearish economic outlook for most of the world
- ESG is now being embraced by as many as 90% of S&P 500 companies
“Classifying data and ensuring that it is stored using the best practices for each classification level (storage media, physical location, destruction method, etc.) to meet the unique business, IT and budgetary requirements will become paramount,” Johnston said.
“While there may continue to be debate about exactly how classification frameworks are defined, there will be little if any argument around the requirement for cost-efficient, cybersecure storage media for strict classification levels.”
As a result, we are likely to see continued development of next-generation storage media to fulfill diverse classification frameworks. That may include the development of media that reduces the upfront acquisition cost, and overall total cost of ownership, while ensuring data is kept active, cybersecure and sustainable.
3. Unstructured data automation
Data classification will be increasingly automated, according to Molly Presley, SVP of marketing, Hammerspace.
With the rise of metadata-driven unstructured data workflows, it will be increasingly possible to classify data based on defined attributes within the metadata.
“Data classification will be able to be done automatically based on customized fields found within the metadata, making it easier for IT and users to remain compliant with their datasets without burdensome manual activity,” Presley said.
4. Data classification for business value
David Wagner, senior research director at Avasant Research, said a companion trend to automation is looking at data classification as more than a security best practice.
“When people think of data classification, they often think of military terms, like top secret, that refer to the sensitivity of the data,” Wagner said.
“However, the primary call of storing large amounts of data is to glean business value from that data.”
Data classification has grown to encompass all types of tagging to make customer data more valuable and reduce data silos. This is not to say that data classification won’t always be a security best practice at heart. But gaining other value from the process can be a value add.
5. Skip the metaverse
There has been a lot of news coverage on the metaverse. Facebook even changed its name to Meta and has invested billions in making it a reality.
But for those in storage and IT, Shiva Nathan, founder and CEO of Onymos, said it will remain hype for some time to come.
“While there might be flashes of jazzy product introductions around metaverse technologies, there will not be any mass adoption or game-changing impact in 2023,” Nathan said.
“These technologies will remain just hype for the foreseeable future until more and more enterprises gain a better understanding of this space and its impact.”
