Interposition-based Big Data Storage Management
IBIS is an interposed big-data I/O scheduler for providing performance differentiation for competing applications’ I/Os in a shared
MapReduce-type big-data system. IBIS is open-source and will be available for downloading soon.
vMoodle: Proxy-based Parallel File System Virtualization and Management
vMoodle is a novel virtual-machine-based online education system. It incorporates cloud-hosted virtual machines (e.g., Amazon EC2 virtual machines) to support a variety of course activities with good scalability and availability, and provide these activities through a typical online learning environment (e.g., Moodle) that teachers and students are already familiar with. In addition, the vMoodle App further exploits the increasing popularity of smart phones and tablets to allow users access vMoodle from anywhere at any time. [Read More]
vPFS: Proxy-based Parallel File System Virtualization and Management
A virtualization-based storage management solution for high-performance computing systems. It is designed to support Quality of Service based storage resource management for applications to achieve their desired performance on today’s and future large-scale computing systems. It is developed in collaboration with researchers from the University of Florida and IBM Research. VPFS is open-source and will be available for downloading soon. [Read More]
FMPC: Fuzzy Modeling Predictive Controller
FMPC: An advanced fuzzy modeling based predictive resource controller for data centers and cloud systems. It supports efficient resource allocations for applications with dynamic and complex behaviors and optimizes resource management globally to meet the applications’ Quality of Service requirements and the entire system’s revenue objective. It is developed in collaboration with researchers from the University of Florida and Intel. FMPCis open-source and will be available for downloading soon. [Read More]
DM-CACHE: Generic Block-level Disk Cache & SSD-Optimized Cloud Storage Cache
A general-purposed and highly-customizable block-level disk caching utility for storage systems. It has been optimized for using emerging solid-state drives (SSD) as caches and supporting scalable virtual machine storage in cloud computing systems. It has been deployed for production use by a leading European cloud service provider, CloudVPS since summer 2012. It has also been extended by other industry companies for various purposes, including Facebook (the Flashcache product), Marvell, Kubisys, and Open-E. Dm-cache is open-source and can be downloaded here. [Read More]
Grid Virtual File System (GVFS)
A user-level virtual file system for grid computing which allows application to transparently access data across a grid system with their desired performance, consistency, security, and fault-tolerance. It has been deployed in the production In-VIGO grid computing system since 2003 and has supported thousands of scientific job executions. GVFS is open-source and can be downloaded here. [Read More]
In-VIGO
A virtualization-based middleware system for grid computing. It pioneered the use of virtualization technologies to offer flexible execution environments for scientific applications and collaborative tools with different requirements, integrating virtualized back-ends, and a web-based interface allowing users to submit requests for application execution or creation of collaborative environments. In-VIGO has been in production use since 2003 and served thousands of scientific jobs from over 600 users across 11 countries. [Read More]
BMI Cyberworkstation
A distributed software system for brain-machine interface research. It supports in vivo brain signal sampling, reliable across-laboratory messaging, parallel brain signal modeling, real-time robotic control, and flexible Web-based management. It has been in production use since 2006 by researchers from different laboratories nationwide, including the University of Florida, the University of Miami, the State University of New York, and the University of California. [Read More]