Resource Allocation for High-Volume Streaming Data in Data Centers. Almost all chip vendors are producing new hardware accelerators by combining several units into a single main-board, and therefore making the execution of parallel and distributed run-time primitives not efficient/scalable. This project aims to develop innovative ways to building incremental and iterative computations over massive data sets in a cluster of heterogeneous systems. This will provide a significant reduction of perfo ....Resource Allocation for High-Volume Streaming Data in Data Centers. Almost all chip vendors are producing new hardware accelerators by combining several units into a single main-board, and therefore making the execution of parallel and distributed run-time primitives not efficient/scalable. This project aims to develop innovative ways to building incremental and iterative computations over massive data sets in a cluster of heterogeneous systems. This will provide a significant reduction of performance bottlenecks when running heavily distributed data-driven applications. Expected outcomes will include resource management algorithms that optimise performance at large scale. The project will benefit many areas, including running stateful iterative stream-based data-analysis applications in data centres. Read moreRead less
Blockchain-Enabled Federated Learning for Secure and Decentralised Learning. This project aims to develop novel blockchain-enabled federated learning techniques for secure and decentralised learning. It addresses an important and urgent machine learning problem, that is, the data useful for training machine learning models are often held by different owners who are not willing to share their data due to privacy concerns, resulting in isolated data islands. The project will result in a set of inn ....Blockchain-Enabled Federated Learning for Secure and Decentralised Learning. This project aims to develop novel blockchain-enabled federated learning techniques for secure and decentralised learning. It addresses an important and urgent machine learning problem, that is, the data useful for training machine learning models are often held by different owners who are not willing to share their data due to privacy concerns, resulting in isolated data islands. The project will result in a set of innovative algorithms that provide solutions to the key challenges in blockchain-enabled federated learning. The expected outcomes of the project will dramatically advance the frontier of machine learning and blockchain research, and have massive social and economic benefits for Australia and international communities.Read moreRead less
Cost-effective App Service Management in Edge Computing Environment. This project aims to deliver a framework and a suite of approaches for cost-effective app service management in the edge computing (EC) environment facilitated by the 5G mobile network. Edge computing offers great promises for rapidly advancing mobile and IoT apps in many active domains in Australia, e.g., self-driving cars, medical services, etc. Using a variety of optimization techniques and game theory, this project attacks ....Cost-effective App Service Management in Edge Computing Environment. This project aims to deliver a framework and a suite of approaches for cost-effective app service management in the edge computing (EC) environment facilitated by the 5G mobile network. Edge computing offers great promises for rapidly advancing mobile and IoT apps in many active domains in Australia, e.g., self-driving cars, medical services, etc. Using a variety of optimization techniques and game theory, this project attacks the new challenges in the deployment, delivery and adaptation of app services in the EC environment. The outcomes of this project will significantly promote new mobile and IoT apps over Australia's 5G mobile network by allowing app vendors to manage their services cost-effectively with ease in the EC environment.Read moreRead less
Adaptive context caching for fast concurrent access in Internet of Things. Context-awareness in Internet of Things (IoT) applications has profound impact on smartness, relevance, adaptability, dependability, performance and flexibility of such applications. This project will address the significant knowledge gap by investigating, proposing and validating a novel adaptive context caching scheme for fast near real-time access in multiple concurrent context queries coming from multiple and diverse ....Adaptive context caching for fast concurrent access in Internet of Things. Context-awareness in Internet of Things (IoT) applications has profound impact on smartness, relevance, adaptability, dependability, performance and flexibility of such applications. This project will address the significant knowledge gap by investigating, proposing and validating a novel adaptive context caching scheme for fast near real-time access in multiple concurrent context queries coming from multiple and diverse IoT applications. The outcome will be a critical component of the IoT context management platform called Context-as-a-Service which is currently under development. The expected benefits will be far ranging and applicable to many domains including intelligent transportation, industrial internet and smart cities..Read moreRead less