Digital Twin Could Help Data Servers Better Predict User Needs
Bad news. Your data’s been evicted.
The good news? It didn’t have to be this way.
If only the server had known you still wanted instant access to your data, it could have saved it some space instead of kicking it to the curb. Moreover, it could make wireless networks as a whole faster and more reliable.
But don’t be too hard on your servers. Similar to how servers in a busy restaurant have to prioritize tables on the fly, data servers are constantly asked to make split decisions.
Systems have to decide “which data packages to store and which data packages can be evicted,” says Yuchen Liu, an assistant professor of computer science at NC State University.
At issue is something called edge caching. Caching — the act of storing data on a server that a system or network thinks users will be using (or re-using) in the near future — allows systems to meet user demands more quickly than having to retrieve the data from its original source.
“Systems can’t put everything in edge caches, and storing too much redundant data on an edge server can slow down the server if the data are using too many computational resources,” Liu explains.
The more accurately a system can predict which data users will want — and “how much data the edge servers should be storing” — the better the system can perform, Liu says.
So Liu set out to investigate ways to help a server more accurately predict which data to store.
Using what’s called a “digital twin” — which effectively clones the network it’s supporting — Liu and his research team have developed a new edge caching optimization method, which they’ve dubbed D-REC.
A computational modeling technique, digital twins are virtual models of a real object. In D-REC’s case, the digital twin is a virtual model of a defined wireless network — either a cellular network or a Wi-Fi network.
“The method can be applied to any wireless network, depending on the system administrator or network operator’s needs,” Liu says. “D-REC can be adjusted depending on the needs of the user.”
Liu is a corresponding author of a paper on the research behind D-REC, which was recently published in the IEEE Journal on Selected Areas in Communications.
Titled “Digital Twin-Assisted Data-Driven Optimization for Reliable Edge Caching in Wireless Networks,” NC State Ph.D. student Zifan Zhang was the first author, and co-authors included Zhiyuan Peng, a postdoctoral researcher at NC State; Dongkuan Xu, an assistant professor of computer science at NC State; Mingzhe Chen of the University of Miami; and Shuguang Cui of the Chinese University of Hong Kong.
This article is based on a news release from NC State University.
