Editing Supercomputer (section)

===Performance metrics===
{{See also|LINPACK benchmarks|Grid computing#Fastest virtual supercomputers}}
[[File:Supercomputing-rmax-graph2.svg|upright=1.5|thumb|Top supercomputer speeds: [[Logarithmic scale|logscale]] speed over 60&nbsp;years]]
In general, the speed of supercomputers is measured and [[Benchmark (computing)|benchmarked]] in [[FLOPS]] (floating-point operations per second), and not in terms of [[Million instructions per second|MIPS]] (million instructions per second), as is the case with general-purpose computers.<ref name=Xifu >{{cite book |title=Performance Evaluation, Prediction and Visualization of Parallel Systems |author= Xingfu Wu |year=1999 |isbn= 978-0-7923-8462-5 |pages= 114–117 |url=https://books.google.com/books?id=IJZt5H6R8OIC&pg=PA116|publisher=Springer Science & Business Media}}</ref> These measurements are commonly used with an [[SI prefix]] such as [[tera-]], combined into the shorthand TFLOPS (10<sup>12</sup> FLOPS, pronounced ''teraflops''), or [[peta-]], combined into the shorthand PFLOPS (10<sup>15</sup> FLOPS, pronounced ''petaflops''.) [[Petascale]] supercomputers can process one quadrillion (10<sup>15</sup>) (1000&nbsp;trillion) FLOPS. [[Exascale computing|Exascale]] is computing performance in the exaFLOPS (EFLOPS) range. An EFLOPS is one quintillion (10<sup>18</sup>) FLOPS (one million TFLOPS). However, The performance of a supercomputer can be severely impacted by fluctuation brought on by elements like system load, network traffic, and concurrent processes, as mentioned by Brehm and Bruhwiler (2015).<ref>Brehm, M. and Bruhwiler,
D. L. (2015) 'Performance Characteristics of the Plasma Wakefield Acceleration Driven by Proton Bunches'. Journal of Physics: Conference Series</ref>

No single number can reflect the overall performance of a computer system, yet the goal of the Linpack benchmark is to approximate how fast the computer solves numerical problems and it is widely used in the industry.<ref name = "Jackpaper">{{Citation
|last1 = Dongarra
|first1 = Jack J.
|last2 = Luszczek
|first2 = Piotr
|last3 = Petitet
|first3 = Antoine
|title = The LINPACK Benchmark: past, present and future
|year = 2003
|journal = Concurrency and Computation: Practice and Experience
|volume = 15
|issue = 9
|pages = 803–820
|url = http://www.netlib.org/utk/people/JackDongarra/PAPERS/hplpaper.pdf
|doi = 10.1002/cpe.728
|s2cid = 1900724
}}</ref> The FLOPS measurement is either quoted based on the theoretical floating point performance of a processor (derived from manufacturer's processor specifications and shown as "Rpeak" in the TOP500 lists), which is generally unachievable when running real workloads, or the achievable throughput, derived from the [[LINPACK benchmarks]] and shown as "Rmax" in the TOP500 list.<ref>{{cite web|title=Understanding measures of supercomputer performance and storage system capacity|url=https://kb.iu.edu/d/apeq#measure-flops|publisher=Indiana University|access-date=3 December 2017}}</ref> The LINPACK benchmark typically performs [[LU decomposition]] of a large matrix.<ref>{{cite web|title=Frequently Asked Questions|url=https://www.top500.org/resources/frequently-asked-questions/|website=TOP500.org|access-date=3 December 2017}}</ref> The LINPACK performance gives some indication of performance for some real-world problems, but does not necessarily match the processing requirements of many other supercomputer workloads, which for example may require more memory bandwidth, or may require better integer computing performance, or may need a high performance I/O system to achieve high levels of performance.<ref name = "Jackpaper"/>