Networking and Algorithms for Big Data
Big Data refers to a collection of information that is too vast and complex to be effectively collected, processed, and analyzed using traditional algorithms, tools, and approaches. In their quest to comprehend and utilize big data, researchers, businesses, and governments are focusing efforts on datasets characterized by three research challenges: volume, velocity, and variety. These challenges requires research and innovation at all levels of computing, from the physical network needed for capturing and transporting such data to advanced algorithms for efficiently and effectively securing, organizing, processing, and, ultimately, making effective use of such data.
The Networks and Algorithms for Big Data is proposed as a Strategic Research Area to address the following problems:
Networks for the Acquisition of Big Data: Challenges in Big Data start with data acquisition, which can be observed in many applications like location-aware sensing in wireless systems, massive MIMO, network traffic monitoring, and collaborative spectrum sensing in cognitive radio systems. The ultimate goal is to acquire and transmit the data in a timely, effective and efficient manner. Research in this area includes, but is not limited to the following three aspects: 1) hierarchical and distributed data filtering and compression, 2) energy efficiency, and 3) data synchronization and network resilience.
Computation for Big Data: Advancements in high performance and cloud computing have made it possible to process a voluminous amount of data from a wide range of sources. The development of faster and cheaper hardware, distributed across multiple nodes, has allowed us to develop and implement complex, high performance parallel algorithms for processing in near real-time.
Knowledge Discovery in Big Data: The heterogeneous, relational and time dependent nature of Big Data has made it possible to analyze data for novel discoveries. The improved capabilities of data mining technologies have allowed us to develop and implement novel statistical, graph-based and hybrid data mining and machine learning approaches for the discovery of interesting patterns and anomalies.
Privacy and Security of Big Data: Privacy and security issues are magnified by all the characteristics of Big Data including volume, velocity, variety and variability. Traditional security mechanisms tailored for small-scale static localized data are not readily portable in such environment. Security and privacy challenges specific to Big Data include (but are not limited to) secure distributed computing, security practices in non-relational data storage, high availability of data and transaction records, input validation in data collection, real-time security/compliance monitoring, efficient access control and secure communication framework and privacy preserving data analytics.
Each of these research topics plays a vital role in the acquisition, processing, and analysis of Big Data in domains such as smart grid, healthcare, earth sciences, resilient infrastructure, cyber-security, and national defense.
- Sheikh Ghafoor
- William (Bill) Eberle
- Sheikh Ghafoor
- Nan (Terry) Guo
- Mohamed Mahmoud
- Robert Qiu
- Stephen Scott
- Ambareen Siraj
- Doug Talbert
- Institute for Modeling, Simulation and Discovery
- Advanced Health Informatics
- Secure Cyberspace
- Restore and Improve Urban Infrastructure
- Engineer the Tools of Scientific Discovery
Current and Potential Partners:
- USF, BYU, U of Ark, UTK, Wright State, ARO, ONR, AFRL, WSU, NSA, UCF, Vanderbilt, DOE, DOD, NSF, ORNL, Sandia and other FFRC’s