Issue No.01 - January/February (2007 vol.27)
Hyesoon Kim , University of Texas at Austin
Onur Mutlu , Microsoft Research
Yale N. Patt , University of Texas at Austin
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MM.2007.9
The branch misprediction penalty is a major performance limiter and a major cause of wasted energy in high-performance processors. The diverge-merge processor reduces this penalty by dynamically predicating a wide range of hard-to-predict branches at runtime in an energy-efficient way that doesn't significantly increase hardware complexity or require major ISA changes.
predication, dynamic predication, branch prediction, adaptivity, pipelining, instruction level parallelism, energy efficiency
Hyesoon Kim, Onur Mutlu, Yale N. Patt, "Diverge-Merge Processor: Generalized and Energy-Efficient Dynamic Predication", IEEE Micro, vol.27, no. 1, pp. 94-104, January/February 2007, doi:10.1109/MM.2007.9