19 January 2023 in ACM Transactions on Storage
Operating systems include many heuristic algorithms designed to improve overall storage performance and throughput. Because such heuristics cannot work well for all conditions and workloads, system designers resorted to exposing numerous tunable parameters to users—thus burdening users with continually optimizing their own storage systems and applications. Storage systems are usually responsible for most latency in I/O-heavy applications, so even a small latency improvement can be significant. Machine learning (ML) techniques promise to learn patterns, generalize from them, and enable optimal solutions that adapt to changing workloads. We propose that ML solutions become a first-class component in OSs and replace manual heuristics to optimize storage systems dynamically. In this article, we describe our proposed ML architecture, called KML. We developed a prototype KML architecture and applied it to two case studies: optimizing readahead and NFS read-size values. Our experiments show that KML consumes less than 4 KB of dynamic kernel memory, has a CPU overhead smaller than 0.2%, and yet can learn patterns and improve I/O throughput by as much as 2.3× and 15× for two case studies—even for complex, never-seen-before, concurrently running mixed workloads on different storage devices.
7 September 2012 in Journal of Assistive Technologies
Purpose: Computer‐based systems for motor function rehabilitation have been around for more than a decade, with work done to help recovery of function in the lower limb (ankle, leg) as well as upper limb (hand and arm).Design/methodology/approach: More recently there has been a trend towards the use of game‐based systems to deliver rehabilitation goals. The authors' interdisciplinary group has been working in the area of motor function recovery of the hand and arm (following stroke) for a number of years, using both high‐end virtual reality (VR) technology as well as low‐cost video capture technology.Findings: Over this time it has become clear that there are many challenges in designing usable, effective game‐based systems for motor function rehabilitation.Originality/value: This paper reflects on user experiences across the range of technologies developed by the group. It presents a summary review of the authors' systems and details the protocols and user evaluation instruments used. It then critically reflects on this work and reviews other recent advances in game usability and playability, leading to suggestions for how the user experience of games for rehabilitation may be improved in future work.