Real-Time Scheduling on Multiprocessors & Non-Blocking Synchronization
◈ Title : Real-Time Scheduling on Multiprocessors & Non-Blocking Synchronization
◈ Speaker : Prof. Hyeonjoong Cho (Korea University)
◈ Date & Time : Friday, June 10, 2011 (14:00pm ~ 15:30pm)
◈ Place : LG Research Building, Room #101
◈ Host : Prof. Se-Young Oh (Tel. 2214)
◈ Abstract : Multiprocessor architectures are becoming more attractive for embedded systems, primarily because major processor manufacturers are making them decreasingly expensive. This makes such architectures very desirable for embedded system applications with high computational workloads, where additional, cost-effective processing capacity is often needed. Responding to this trend, RTOS vendors are increasingly providing multiprocessor platform support. But this exposes the critical need for real-time scheduling and synchronization for multiprocessors — a comparatively undeveloped area of real-time scheduling which has recently received significant research attention.
We will discuss about an optimal real-time scheduling algorithm for multiprocessors, one that satisfies all task deadlines, when the total utilization demand does not exceed the utilization capacity of the processors. The algorithm, called LLREF, is designed based on a novel abstraction for reasoning about task execution behavior on multiprocessors: the Time and Local Execution Time Domain Plane (or T-L plane). We will also discuss about a wait-free protocol for the single-writer/multiple-reader problem in small-memory embedded real-time systems. We present an analytical solution to obtain the minimum (optimal) space cost required for this problem considering a-priori knowledge of interferences—the first established.