Tornado warning: The perils of selective replay in multithreaded processors

Yongxiang Liu*, Anahita Shayesteh, Gokhan Memik, Glenn Reinman

*Corresponding author for this work

Research output: Contribution to conferencePaperpeer-review

2 Scopus citations


As future technologies push towards higher clock rates, traditional scheduling techniques that are based on wake-up and select from an instruction window fail to scale due to their circuit complexities. Speculative instruction schedulers can significantly reduce logic on the critical scheduling path, but can suffer from instruction misscheduling that can result in wasted issue opportunities. Misscheduled instructions can spawn other misscheduled instructions, only to be replayed over again and again until correctly scheduled. These "tornadoes" in the speculative scheduler are characterized by extremely low useful scheduling through-put and a high volume of wasted issue opportunities. The impact of tornadoes becomes even more severe when using Simultaneous Multithreading. Misschedulings from one thread can occupy a significant portion of the processor issue bandwidth, effectively starving other threads. In this paper, we propose Zephyr, an architecture that inhibits the formation of tornadoes. Zephyr makes use of existing load latency prediction techniques as well as coarse-grain FIFO queues to buffer instructions before entering scheduling queues. On average, we observe a 23% improvement in IPC performance, 60% reduction in hazards, 41% reduction in occupancy, and 48% reduction in the number of replays compared with a baseline scheduler.

Original languageEnglish (US)
Number of pages10
StatePublished - Dec 1 2005
EventICS05 - 19th ACM International Conference on Supercomputing - Cambridge, MA, United States
Duration: Jun 20 2005Jun 22 2005


OtherICS05 - 19th ACM International Conference on Supercomputing
CountryUnited States
CityCambridge, MA

ASJC Scopus subject areas

  • Computer Science(all)

Fingerprint Dive into the research topics of 'Tornado warning: The perils of selective replay in multithreaded processors'. Together they form a unique fingerprint.

Cite this