Summary
Total value of grants received:
- As the PI: SGD 300,000
- As a Co-I: allocated SGD 534,000 out of SGD 2,184,482
Handling Terabytes of New Data Each Day
Rolls Royce Master Research Grant (Principal Investigator), August 2010 - August 2011, SGD 100,000
Project description: The aerospace industry is shifting towards a more service-based business model from a traditional pure original equipment manufacturer (OEM) approach. In this context, as the demands of commercial as well as military aviation soar, the requirements on service quality of the aerospace aftermarket businesses continue to become more stringent. Particular challenges exist in the area of equipment health monitoring (EHM) applications through which aftermarket businesses offer round-the-clock support to their customers through the monitoring and servicing of in-service products. This project investigates the following issues related to EHM applications:
- Effective data analytics and process mining for EHM with rapidly increasing data volumes.
- Efficient access to dynamically generated data for running analytics in real-time to ensure maximum service quality.
- Rapid validation of new analytical methods on massive volumes of historical data.
- Effective monitoring of services for increasing number of in-service products coupled with increasing sensing devices.
Future Data Center Technologies - Large Scale Hybrid Storage System
A*STAR Thematic Strategic Research Programme (Co-Investigator), August 2011 - August 2014, SGD 1,286,210.24
Project description: The world is moving rapidly toward a service-oriented economy. Effective data analytics enables services to be personalizable, predictive, proactive and adaptive. However, the data to be collected, managed and analyzed by the service industry is getting too massive for many enterprises to handle economically and efficiently with its own resources (both hardware and software). In this project, we propose a Cloud-based service platform to provide enterprises with IT functions over the Internet. Our objective is to realize the next generation data analytics by developing user and domain-driven data analytics as a service on the Cloud. We aim at developing a holistic scalable distributed data analytics framework that complements users of analytics by extraction, integration,
summarization and visualization of relevant heterogeneous information from distributed sources, and facilitates user interpretation, interaction and collaboration to solve domain-specific problems.
Future Data Center Technologies - Large Scale Hybrid Storage System
A*STAR Thematic Strategic Research Programme (Co-Investigator), August 2011 - August 2014, SGD 898,272
Project description: Many business applications today demand massively scalable and ultra-efficient backend storage support, e.g., high capacity (in the range of petabytes), high IO throughput (greater than 200 GB/sec), and large IOP (input/output per second). To cope up with these demanding workloads, this project proposes to develop a very high performance hybrid petabyte-scale storage system based on appliances utilizing non-volatile memory (NVM), solid-state drive (SSD), hard disk drive, and hybrid drives that are primarily built with commodity hardwares and operating systems. The project involves efficient metadata management, proactive data migration based on workload monitoring and prediction, and novel data placement and migration techniques for handling enterprise application workloads. The project also investigates the potential of building a next-generation data center to provide enterprises with comprehensive backend data storage services on the Cloud.
Correlation Discovery in Graph Databases and Its Applications
ACRF Tier 1 Grant (Principal Investigator), March 2010 - February 2012, SGD 100,000
Project description: Graph exists ubiquitously in a wide spectrum of application domains, such as social networks, mobile and telecommunication networks, financial networks, marketing networks, Web graphs, P2P networks, road networks, bio-informatics networks, etc. This project studies the inter-relationship between the hidden patterns in these complex networks. The project also aims at applying the discovered knowledge to develop new solutions for applications in information systems, particularly those in telecommunication, social media and Infocomm applications.
Large-Scale Graph Data Management
NTU Start-Up Grant (Principal Investigator), June 2009 - May 2012, SGD 100,000
Project description: This project focuses on designing and developing efficient large-scale graph data management technologies including graph data mining, graph pattern matching and retrieval, graph indexing, and graph data storage, with applications in corporate management, anomaly detection, user behavioral analysis, schema matching and integration, pattern recognition, law enforcement, computer vision, graph-motif discovery, etc.