Carding Forum
Professional
- Messages
- 2,788
- Reaction score
- 1,198
- Points
- 113
The new caching method improves performance and reliability.
Computer science scientists have developed a method for predicting the data needed by wireless network users before they need it, making networks faster and more reliable. This method uses a technique called "digital twinning", which effectively clones the supported network.
The method is based on the concept of caching at the edge of the network. Caching is storing data on a server that the system expects users to use in the near future. This allows the system to meet the needs of users faster than when extracting data from the original source. Network edge caching means storing data on a server that is located closest to the end user, such as on computers integrated into or located near network routers.
"The two main tasks here are to determine what data needs to be cached, and how much data should be stored by the server at the edge of the network at any given time," says Yuchen Liu, co-author of the study and an assistant professor of computer science at North Carolina State University.
Systems cannot store all data in caches at the edge of the network, and excessive storage can slow down the server if the data uses too much computing resources. As a result, systems are constantly making decisions about which data packets to store and which to delete.
"The more accurately the system predicts what data users will need and how much data servers at the edge of the network should store, the better the system will perform," Liu explains. "Our work has focused on improving these predictions."
A new method for optimizing caching at the edge of the network, called D-REC, uses a computational simulation called a digital double. A digital double is a virtual model of a real object. In the case of D-REC, a digital twin is a virtual model of a particular wireless network, whether it's a cellular network or Wi-Fi.
"The method can be applied to any wireless network, depending on the needs of the system administrator or network operator, "Liu notes." D-REC can be customized according to the user's needs."
In D-REC, the digital twin receives real-time data from the wireless network and uses it to run simulations, predicting what data is most likely to be requested by users. These predictions are then passed back to the network to make caching decisions at the edge of the network. Since simulations are performed by a computer outside the network, this does not slow down the network itself.
The researchers used open data sets to determine whether the wireless network performed more efficiently with D-REC. They conducted extensive experiments that took into account many variables, such as network scale, number of users, and others.
"D-REC has surpassed traditional approaches," says Liu. "Our method has improved the network's ability to accurately predict what data should be cached at the edge. D-REC also helped systems better balance data storage across the entire network."
In addition, since the D-REC digital twin focuses on predicting network behavior, it can identify potential problems in advance.
"For example, if a digital doppelganger believes that there is a high probability of overloading a particular base station or server, the network can be notified of this, which will allow data to be redistributed across the network to preserve performance and reliability," explains Liu.
"For now, we are ready to work with network operators to explore how D-REC can improve network performance and reliability in real-world environments."
The paper "Digital Twin-Assisted Data-Driven Optimization for Reliable Edge Caching in Wireless Networks" was published in the IEEE Journal on Selected Areas in Communications. The first author of the paper is Qifan Zhang, a graduate student at North Carolina State University. Co-authors also included Zhiyuan Peng, a postdoctoral fellow at the same university, Dongkuan Xu, an associate professor of Computer Science, Mingzhe Chen of the University of Miami, and Shuguang Cui of the Chinese University of Hong Kong.
Source
Computer science scientists have developed a method for predicting the data needed by wireless network users before they need it, making networks faster and more reliable. This method uses a technique called "digital twinning", which effectively clones the supported network.
The method is based on the concept of caching at the edge of the network. Caching is storing data on a server that the system expects users to use in the near future. This allows the system to meet the needs of users faster than when extracting data from the original source. Network edge caching means storing data on a server that is located closest to the end user, such as on computers integrated into or located near network routers.
"The two main tasks here are to determine what data needs to be cached, and how much data should be stored by the server at the edge of the network at any given time," says Yuchen Liu, co-author of the study and an assistant professor of computer science at North Carolina State University.
Systems cannot store all data in caches at the edge of the network, and excessive storage can slow down the server if the data uses too much computing resources. As a result, systems are constantly making decisions about which data packets to store and which to delete.
"The more accurately the system predicts what data users will need and how much data servers at the edge of the network should store, the better the system will perform," Liu explains. "Our work has focused on improving these predictions."
A new method for optimizing caching at the edge of the network, called D-REC, uses a computational simulation called a digital double. A digital double is a virtual model of a real object. In the case of D-REC, a digital twin is a virtual model of a particular wireless network, whether it's a cellular network or Wi-Fi.
"The method can be applied to any wireless network, depending on the needs of the system administrator or network operator, "Liu notes." D-REC can be customized according to the user's needs."
In D-REC, the digital twin receives real-time data from the wireless network and uses it to run simulations, predicting what data is most likely to be requested by users. These predictions are then passed back to the network to make caching decisions at the edge of the network. Since simulations are performed by a computer outside the network, this does not slow down the network itself.
The researchers used open data sets to determine whether the wireless network performed more efficiently with D-REC. They conducted extensive experiments that took into account many variables, such as network scale, number of users, and others.
"D-REC has surpassed traditional approaches," says Liu. "Our method has improved the network's ability to accurately predict what data should be cached at the edge. D-REC also helped systems better balance data storage across the entire network."
In addition, since the D-REC digital twin focuses on predicting network behavior, it can identify potential problems in advance.
"For example, if a digital doppelganger believes that there is a high probability of overloading a particular base station or server, the network can be notified of this, which will allow data to be redistributed across the network to preserve performance and reliability," explains Liu.
"For now, we are ready to work with network operators to explore how D-REC can improve network performance and reliability in real-world environments."
The paper "Digital Twin-Assisted Data-Driven Optimization for Reliable Edge Caching in Wireless Networks" was published in the IEEE Journal on Selected Areas in Communications. The first author of the paper is Qifan Zhang, a graduate student at North Carolina State University. Co-authors also included Zhiyuan Peng, a postdoctoral fellow at the same university, Dongkuan Xu, an associate professor of Computer Science, Mingzhe Chen of the University of Miami, and Shuguang Cui of the Chinese University of Hong Kong.
Source
