CPU性能评估标准

原文地址: http://blog.chinaunix.net/u/21948/showart_466493.html

---

MIPS： Million Instructions executed Per Second

DMIPS： Dhrystone Million Instructions executed Per Second

MFLOPS： Million Floating-point Operations Per Second

 

    MIPS，DMIPS，MFLOPS是常用的CPU性能评估标准。注意MIPS只适合评估标量机，而MFLOPS比较适用于衡量向量机的性能。

 

    用DMIPS来衡量CPU性能存在很大的缺陷。DMIPS并非字面上每秒百万条指令的意思，而是运行一个叫Dhrystone的基准程序时表现出来的相对性能高低的单位，很多场合人们也习惯用MIPS作为这个性能指标的单位。因为基于程序的测试容易收到恶意优化的干扰，并且DMIPS指标值的发布不受任何机构的监督，所以使用DMIPS进行评估时要慎重。ARM一直采用比较保守的值作为CPU的DMIPS的标称值。DMIPS另外一个缺点是不能测量处理器的数字信号处理能力和Cache/MMU子系统的性能。因为Dhrystone测试程序不包含DSP表达式，只包含一些整型运算和字符串处理，并且测试程序偏小，几乎可以完整地放在Cache里面运行而无需与外部存储器进行交互。这样就难以反映处理器在一个真实系统中的真正性能。

 

    资料：

Dhrystone is a synthetic benchmark program developed in 1984 by Reinhold P. Weicker intended to be representative of system (integer) programming. The Dhrystone grew to become representative of general processor (CPU) performance until it was superseded by the CPU89 benchmark suite from the Standard Performance Evaluation Corporation, today known as the "SPECint" suite.With Dhrystone, Weicker gathered meta-data from a broad range of software - including programs written in FORTRAN, PL/1, SAL, ALGOL 68, and Pascal. He then characterized these programs in terms of various common constructs - procedure calls, pointer indirections, assignments, etc. From this he wrote the Dhrystone benchmark to correspond to a representative mix. Dhrystone was published in Ada, with the C version for Unix developed by Rick Richardson ("version 1.1") greatly contributing to its popularity.

Dhrystone vs Whetstone

The Dhrystone benchmark contains no floating point operations, thus the name is a pun on the then-popular Whetstone benchmark for floating point operations. The output from the benchmark is the number of Dhrystones per second (the number of iterations of the main code loop per second).

Both Whetstone and Dhrystone are synthetic benchmarks, meaning that they are simple programs that are carefully designed to statistically mimic some common set of programs. Whetstone, developed in 1972, originally strove to mimic typical Algol 60 programs based on measurements from 1970, but eventually became most popular in its Fortran version. Whetstone thus reflected the highly numerical orientation of computing in the 1960s.

Issues addressed by Dhrystone

Dhrystone's eventual importance as an indicator of general-purpose ("integer") performance of new computers made it a target for commercial compiler writers. Various modern compiler techniques (such as dead code elimination) make the use and design of synthetic benchmarks more difficult. Version 2.0 of the benchmark, released by Weicker and Richardson in March of 1988, had a number of changes intended to foil a range of compiler techniques. Yet it was carefully crafted so as not to change the underlying benchmark. This effort to foil compilers was only partly successful. Dhrystone 2.1, released in May of the same year, had some minor changes and remains the current definition of Dhrystone.

Other than issues related to compiler optimization, various other issues have been cited with the Dhrystone. Most of these were understood at the time of its publication in 1984 - including the small code size and small data set size. More subtle is the slight over-representation of string operations, which is largely language related: both Ada and Pascal have strings as first class citizens in the language, whereas C does not, so what was simple variable assignments in reference benchmarks became buffer copy operations in the C library.

Dhrystone remains remarkably resilient as a simple benchmark. It is easy to use, well documented, is fully self-contained, is well understood, and can be made to work on almost any system. In particular, it has remained in broad use in the embedded computing world, though the recently developed EEMBC benchmark suite as well as HINT, Stream, and even Bytemark are widely quoted and used, as well as more specific benchmarks for memory subsystem (Cachebench), TCP/IP (TTCP), and many others. Still, 20 years of continued use is quite a testament to Weicker's careful design and foresight.

Results

Dhrystone tries to represent the result more meaningfully than MIPS (million instructions per second), because MIPS cannot be used across different instruction sets (e.g. RISC vs. CISC) for the same computation requirement from users. Thus, the main score is just Dhrystone loops per second. Another common representation of the Dhrystone benchmark is the DMIPS - Dhrystone MIPS - obtained when the Dhrystone score is divided by 1,757 (the number of Dhrystones per second obtained on the VAX 11/780, nominally a 1 MIPS machine).
摘自wiki


 

另外自己也补充一点

Whetstones

Whetstone is a benchmark test which attempts to measure the speed and efficiency at which a computer performs floating-point operations. The result of the test is given in units called kilo-whetstones-per-second or KWIPS.

The Whetstone is a synthetic benchmark designed to measure the behavior of scientific programs. It contains several modules that are meant to represent a mix of operations typically performed in scientific applications. A wide variety of C functions including sin, cos, sqrt, exp, and log are used as well as integer and floating-point math operations, array accesses, conditional branches, and procedure calls. The primary aim of this benchmark is to measure the performance of both integer and floating-point arithmetic.

Dhrystones

Dhrystone is a general-performance benchmark test originally developed by Reinhold Weicker in 1984. This benchmark is used to measure and compare the performance of different computers or, in this case, the efficiency of the code generated for the same computer by different compilers. The test reports general performance in Dhrystone per second.

Like most benchmark programs, Dhrystone consists of standard code and concentrates on string handling. It uses no floating-point operations. It is heavily influenced by hardware and software design, compiler and linker options, code optimizing, cache memory, wait states, and integer data types.

from http://www.keil.com/benchmarks/