Data is growing exponentially in every organization, but large school systems at the primary and secondary levels face their own set of challenges when it comes to managing, securing and better utilizing information.
For Mark Lamson, director of information technology at Westerly Public Schools, the task is to manage data sprawl created by more than 3,000 students and nearly 700 faculty and staff that are coming and going semester to semester. Managing and storing data while securing it and remaining compliant with rules and regulations that govern the institution are critical, as with any enterprise. There is also a desire to improve the educational experience by using data to make better decisions.
In a recent webinar hosted by DataGravity, Three Greatest Challenges for Data Management: IT Director’s Perspective, Lamson said unstructured data has grown at a rate of 60 to 80 per cent over the last 15 years, and now has 160TB to contend with. “We’re no stranger to the idea that data is going to keep growing.”
Until recently, the protocol at Westerly was that if in doubt, keep everything because you never know what you might need later. “That’s not a well-advised strategy,” Lamson said. For one thing, it’s not sustainable given the cost of on-premise storage, and it’s not an effective use of that storage. And that data must be secured to meet regulatory and compliance guidelines, even if it’s no longer accessed or useful. “In some cases, keeping stuff has become more of a liability.”
David Stevens, technical marketing manager at DataGravity, worked in IT at a university in a past life. It had a “keep all” data policy and similar data growth rate, but budgets and resources that were shrinking. Because there was so much churn among students and staff, even more than a K12 school or board, it was difficult to keep data growth in check. Best practices needed to be developed to understand all of the data in the organization, including dormant data that hadn’t been touched in more than a semester, so that it could be archived or moved to cold storage.
No matter the age of the data or its sensitivity, it must be secured, and since it’s not a question of when a data breach will occur, but when, Lamson said he focuses on the “crown jewels,” which is data with personally identifiable information (PII). That includes information that might identify a student or staff member, as well as performance data, and how it is accessed, captured and shared, given that breaches are often the result of inappropriate handling and use of data by staff.
Lamson is able to quickly search his data and to understand who might have access to it. For example, it might just be a payroll officer, and he can verify the information was not inappropriately shared or otherwise exposed.
The growth of data and keeping it secure and compliant has meant that service level agreements (SLAs) have evolved, said Lamson. Rather than being about uptime and availability – the five nines – they revolve around data governance and the five W’s of data.
In a university setting, understanding who has access to what data and how they are sharing it is even more complex, said Stevens. “Every semester we have a new group of hackers coming to campus.” The more frequent churn and the variety of data, ranging from student records to research, meant it was necessary to develop guidelines for data classification and protection. “Once they were in place we were able to push it across the organization and help departments understand what they have to do in case of a breach,” he said.
An end game for educational organizations is that once the data is properly managed, secure and compliant, then it can be used to make decisions that will improve the delivery of education and address issues affecting students and faculty, Lamson said. “It’s fairly easy if you have a traditional, structured database that spits out the information you want.”
With unstructured data the challenge is not only finding the needle in the haystack, but finding the haystack. But with the right hardware and software combination, it’s possible to drill down into dark and murky data, Lamson said. For example, issues such as truancy in older students could be nipped in the bud if found early on with the right tool set. It’s also easier to decide what data can be deleted completely with confidence given the cost of storing it and remaining compliant.
At a university level, said Stevens, where information is scattered across campus, bringing it together meant better insights and how to utilize it more. “By bringing together data, you can understand how people are using it and maintain governance across the organization.”
The ability to better understand data residing within a large organization and make better decisions speaks to the growth of what research firm Taneja Group describes as “data aware storage,” of which DataGravity is one of the vendors.