Home Bio Publications Contact

An In-Depth Look at Computer Performance Growth
Authors: Magnus Ekman, Fredrik Warg and Jim Nilsson
Abstract:

It is a common belief that computer performance growth is over 50\% annually, or that performance doubles every 18-20 months. By analyzing publicly available results from the SPEC integer (CINT) benchmark suites, we conclude that this was true between 1985 and 1996 the early years of the RISC paradigm. During the last 7.5 years (1996-2004), however, performance growth has slowed down to 41\%, with signs of a continuing decline. Meanwhile, clock frequency has improved with about 29\% annually. The improvement in clock frequency was enabled both by an annual device speed scaling of 20\% as well as by longer pipelines with a lower gate-depth in each stage. This paper takes a fresh look at and tries to remove the confusion about performance scaling that exists in the computer architecture community.

Keywords: Performance evaluation
Year-Month: 2005-03
Published: ACM SIGARCH Computer Architecture News
Publication type: Newsletter
Vol: 33 No: 1 Pages: 144--147
Bibtex:
@article{PerfGrowth_canews2005,
  title     = {An In-Depth Look at Computer Performance Growth},
  author    = {Ekman, Magnus and Warg, Fredrik and Nilsson, Jim},
  year      = {2005},
  month     = {03},
  abstract  = {It is a common belief that computer performance growth is over 50\% annually, or that performance doubles every 18-20 months. By analyzing publicly available results from the SPEC integer (CINT) benchmark suites, we conclude that this was true between 1985 and 1996 the early years of the RISC paradigm. 

During the last 7.5 years (1996-2004), however, performance growth has slowed down to 41\%, with signs of a continuing decline. Meanwhile, clock frequency has improved with about 29\% annually. The improvement in clock frequency was enabled both by an annual device speed scaling of 20\% as well as by longer pipelines with a lower gate-depth in each stage. This paper takes a fresh look at and tries to remove the confusion about performance scaling that exists in the computer architecture community.},
  keywords  = {Performance evaluation},
  journal   = {{ACM SIGARCH} Computer Architecture News},
  doi       = {10.1145/1055626.1055646},
  volume    = {33},
  number    = {1},
  pages     = {144--147},
  note      = {Publication data: https://warg.org/fredrik/publ/}
}