News
Search All News
Monday, May 4, 2026
Office Hours
Monday - Friday 9am - 5pmClosed Sat. and Sun.
What Happens When Data Needs a Memory?
May 4, 2026
A new paper published in Informatics by Assistant Professor of Finance, Information Systems, and Economics Di Wu brings a powerful, but until now largely theoretical, approach to data storage closer to real-world use. The model Wu studies, called BiTRDF, is designed to keep a reliable, layered record of how information changes over time.
In medical records, for instance, if a patient’s diagnosis changes, the old diagnosis should not disappear from the chart. It should show what was recorded at the time, what changed later, and when it changed. Most databases are better at one part of that task than the other. BiTRDF is designed to handle it all.
To see how well the model could work in practice, Wu and his co-authors used a common computer language called Python to build and testsix versions of the system using datasets that ranged from hundreds of thousands to tens of millions of entries. All six performed well with smaller datasets. At larger scales, some handled the added volume more efficiently, while others were easier to modify or extend.
The study shows that the approach can be built with widely available tools, and gives researchers and developers a clearer sense of which version may work best for different needs. Potential uses include financial records, historical research, health care, and any field where it is important to know not only what changed, but when and how it changed.
Read the paper here.