M
Matthew C. Merten
Researcher at Intel
Publications - 62
Citations - 1072
Matthew C. Merten is an academic researcher from Intel. The author has contributed to research in topics: Reservation station & Instruction register. The author has an hindex of 18, co-authored 62 publications receiving 1063 citations. Previous affiliations of Matthew C. Merten include University of Illinois at Urbana–Champaign.
Papers
More filters
Proceedings ArticleDOI
A hardware-driven profiling scheme for identifying program hot spots to support runtime optimization
Matthew C. Merten,Andrew R. Trick,Christopher Neith George,John C. Gyllenhaal,Wen-mei W. Hwu +4 more
TL;DR: A novel hardware-based approach for identifying, profiling, and monitoring hot spots in order to support runtime optimization of general purpose programs using a set of tightly coupled hardware tables and control logic modules placed in the retirement stage of a processor pipeline removed from the critical path is presented.
Proceedings ArticleDOI
Run-time spatial locality detection and optimization
TL;DR: A microarchitecture scheme which detects and adapts to varying spatial locality, dynamically adjusting the amount of data fetched on a cache miss is presented, showing significant speedups in detailed simulations of several integer programs.
Proceedings ArticleDOI
5.9 Haswell: A family of IA 22nm processors
Nasser A. Kurd,Muntaquim Chowdhury,Edward A. Burton,Thomas P. Thomas,Christopher P. Mozak,Brent R. Boswell,Manoj B. Lal,Anant Deval,Jonathan P. Douglas,Mahmoud Elassal,Nalamalpu Ankireddy,Timothy M. Wilson,Matthew C. Merten,Srinivas Chennupaty,Gomes Wilfred,Raghavan Kumar +15 more
TL;DR: The primary goals for the Haswell program are platform integration and low power to enable smaller form factors and an Intel AVX2 instruction set that supports floating-point multiply-add (FMA), and 256b SIMD integer achieving 2× the number of floating- point and integer operations over its predecessor.
Patent
Compare and exchange operation using sleep-wakeup mechanism
TL;DR: In this paper, a sleep-wakeup mechanism is used for compare and exchange operations using an instruction at a processor to help acquire a lock on behalf of the processor, and then the instruction is put to sleep until an event has occurred.
Patent
Method and apparatus for instruction execution hot spot detection and monitoring in a data processing unit
Wen-mei W. Hwu,Matthew C. Merten,Andrew R. Trick,Christopher Neith George,John C. Gyllenhaal +4 more
TL;DR: In this paper, a hot spot detection counter is used to track non-hot spot branches and hot spot candidate branches and when hot-spot candidate branches are frequently encountered compared to nonhot-spots candidate branches, the hot spot detector may notify the operating system and hot-spaces candidate branch addresses may be supplied to a runtime optimizing compiler and a monitor table or a hotspot monitor.