One small step for a man
One Giant leap for the mankind

There is no wealth like Knowledge
                            No Poverty like Ignorance
Journal of Emerging Trends in Computing and Information Sciences Logo

Journal of Emerging Trends in Computing and Information Sciences >> Call for Papers Vol. 8 No. 3, March 2017

Journal of Emerging Trends in Computing and Information Sciences

Prioritizing Write Buffer Occupancy in Simultaneous Multithreading Processors

Full Text Pdf Pdf
Author Sherifdeen Lawal, Yilin Zhang, Wei-Ming Li
ISSN 2079-8407
On Pages 515-522
Volume No. 6
Issue No. 10
Issue Date November 1, 2015
Publishing Date November 1, 2015
Keywords Simultaneous Multi-threading; Superscalar; Write Buffer


Throughput improvement in Simultaneous Multithreading processors is achievable only when critical shared resources among temporally competing threads are ef?ciently controlled. Allowing any of the threads to overwhelm these resources not only leads to unfair thread processing, but also may severely degrade overall system performance. Write buffer in SMT processor is one of the most critical shared resources and lack of balanced utilization of this buffer creates not only a bottleneck but also an irrefutable degradation to the entire system. Furthermore, due to its size constraint and potentially long occupancy latency of its data, it is a crucial component of the SMT processor in need of control and relief from pressure. The commonly adopted algorithm at the commit stage of the SMT processor concentrates on equal opportunity for all threads to retire their instructions (round robin). However, it provides no control over how individual threads retire or commit store instructions to a write-buffer. In this paper, we propose a new prioritized entry algorithm into the write buffer by giving threads with fewer entries already in the buffer higher priorities. Extensive simulation runs show that it can enhance the overall system throughput by an appreciable margin. An improvement in IPC of up to 28% and 129% is observed when the proposed technique is applied to a 4-threaded and 8-threaded SMT system, respectively.

    Journal of Computing | Call for Papers (CFP) | Journal Blog | Journal of Systems and Software | ARPN Journal of Science and Technology | International Journal of Health and Medical Sciences | International Journal of Economics, Finance and Management     
© 2015 Journal of Computing