Microarchitectures of modern processors
Speaker: Per Stenstrom – Gothenburg, SwedenTopic(s): Hardware, Power and Energy , Architecture, Embedded Systems and Electronics, Robotics
Abstract
Exploiting parallelism at all levels of the compute stack is a prerequisite for delivering high compute performance. This ranges from instruction-level parallelism (ILP) to program or thread-level parallelism (TLP). ILP has been a key enabler for high single-thread compute performance. This lecture focuses on that.
This lecture provides an overview of the microarchitecture of modern processors. It starts out with the challenges related to instruction pipelining (hazards and exception handling). It then focuses on concepts for ILP exploitation including out-of-order instruction execution covering concepts such as the Tomasulo algorithm, branch prediction, speculation and multithreaded execution.
About this Lecture
Number of Slides: 25Duration: 45 minutes
Languages Available: English
Last Updated:
Request this Lecture
To request this particular lecture, please complete this online form.
Request a Tour
To request a tour with this speaker, please complete this online form.
All requests will be sent to ACM headquarters for review.