Shulou(Shulou.com)06/02 Report--

There are a lot of articles about how to use UnixBench, so I won't expand it here. Here is a key description of how it is scored.

Operation parameters

Encounter a lot of customers, install, directly. / Run, run out the results, and then only take the last score, who is higher and who is lower.

Here is a list of the results of 4C8G:

-Benchmark Run: June 25 2018 20:25:47-20 54 CPUs in system Running 1 parallel copy of testsDhrystone 2 using register variables 30971628.9 lps (10.0s, 7 samples) Double-Precision Whetstone 3928.1 MWIPS (12.4s, 7 samples) Execl Throughput 3117.6 lps (30.0s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 645027.2 KBps (30.0s, 2 samples) File Copy 256bufsize 500 maxblocks 229505.4 KBps (30.0s 2 samples) File Copy 4096 bufsize 8000 maxblocks 1288742.6 KBps (30.0s, 2 samples) Pipe Throughput 1635960.9 lps (10.0s, 7 samples) Pipe-based Context Switching 337333.8 lps (10.0s, 7 samples) Process Creation 8238.2 lps (30.0s 2 samples) Shell Scripts (1 concurrent) 5817.0 lpm (60.0s, 2 samples) Shell Scripts (8 concurrent) 2183.9 lpm (60.0s, 2 samples) System Call Overhead 2465754.7 lps (10.0s) 7 samples) System Benchmarks Index Values BASELINE RESULT INDEXDhrystone 2 using register variables 116700.0 30971628.9 2654.0Double-Precision Whetstone 55.0 3928.1 714.2Execl Throughput 43.0 3117.6 725.0File Copy 1024 bufsize 2000 maxblocks 3960.0 645027.2 1628.9File Copy 256 Bufsize 500 maxblocks 1655.0 229505.4 1386.7File Copy 4096 bufsize 8000 maxblocks 5800.0 1288742.6 2222.0Pipe Throughput 12440.0 1635960.9 1315.1Pipe-based Context Switching 4000.0 337333.8 843.3Process Creation 126.0 8238.2 653. 8Shell Scripts (1 concurrent) 42.4 5817.0 1371.9Shell Scripts (8 concurrent) 6.02183.9 3639.9System Call Overhead 15000.0 2465754.7 1643.8 = System Benchmarks Index Score 1362.9----Benchmark Run: 25 2018 20:54:19-21 June 22 1362.9----Benchmark Run CPUs in system Running 4 parallel copies of testsDhrystone 2 using register variables 114984418.6 lps (10.0s, 7 samples) Double-Precision Whetstone 16614.2 MWIPS (11.6s, 7 samples) Execl Throughput 13645.3 lps (30.0s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 751698.4 KBps (30.0s, 2 samples) File Copy 256bufsize maxblocks 230211.7 KBps (30.0s 2 samples) File Copy 4096 bufsize 8000 maxblocks 1964420.6 KBps (30.0s, 2 samples) Pipe Throughput 5999380.0 lps (10.0s, 7 samples) Pipe-based Context Switching 1095000.2 lps (10.0s, 7 samples) Process Creation 34454.9 lps (30.0s 2 samples) Shell Scripts (1 concurrent) 18218.1 lpm (60.0s, 2 samples) Shell Scripts (8 concurrent) 2493.1 lpm (60.0s, 2 samples) System Call Overhead 5643267.3 lps (10.0s) 7 samples) System Benchmarks Index Values BASELINE RESULT INDEXDhrystone 2 using register variables 116700.0 114984418.6 9853.0Double-Precision Whetstone 55.0 16614.2 3020.8Execl Throughput 43.0 13645.3 3173.3File Copy 1024 bufsize 2000 maxblocks 3960.0 751698.4 1898.2File Copy 256 bufsize 500 maxblocks 1655.0 230211.7 1391.0File Copy 4096 bufsize 8000 maxblocks 5800.0 1964420.6 3386.9Pipe Throughput 12440.0 5999380.0 4822.7Pipe-based Context Switching 4000.0 1095000.2 2737.5Process Creation 126.0 34454.9 2734.5Shell Scripts (1 concurrent ) 42.4 18218.1 4296.7Shell Scripts (8 concurrent) 6.0 2493.1 4155.1System Call Overhead 15000.0 5643267.3 3762.2 = System Benchmarks Index Score 3357.0

Many people compare it with a score of 3357, but the above single-process result of 1362.9 is also meaningful, but no one cares about so many cases. The result of multiple processes is the result of multiple processes at the same time according to the number of CPU cores of the system; a single process is the result of only one process, and both results are meaningful for system performance.

Then again, what exactly did Run do? The Index module is executed by default, which contains the following Mini Program:

"dhry2reg", "whetstone-double", "execl", "fstime", "fsbuffer", "fsdisk", "pipe", "context1", "spawn", "shell1", "shell8", "syscall"

It corresponds to the case of each result of the above result.

Looking at its source code, its execution parameters are as follows:. / Run module-I iterations-c process concurrency-Q Maximv (output mode.

Module: sometimes you only want to test the results of a Mini Program, then you can modify yourself to add a module, modify the test content, which is very convenient for debugging, you can also directly specify an execution program.

Number of iterations: with regard to the number of iterations, it can not be taken for granted that this is the number of iterations to be executed. The default framework is divided into two iteration modes, one is the number of short iterations, the default is 3, and the other is the number of long iterations, the default is 10. Then the input number of iterations is the default number of long iterations, short iterations = (iterations + 1) / 3, if the result is less than 1. When the program is configured, which number of iterations should be chosen.

Process concurrency: if you want the process concurrency to be not the system CPU number, you can adjust it through this

Output mode: silent or detailed output mode

Count points

Each Mini Program needs to specify the iteration mode, execution time and concurrency when it is executed, and the result of all processes executed at one time is a unified mode.

Single process scores:

Here is an example of a single process:

Process Creation-- 1 copy== > "/ opt/unixbench/UnixBench/pgms/spawn" 302 > & 1 > > "/ opt/unixbench/UnixBench/results/VM_0_13_centos-2018-06-25-05.log" # Pass approval COUNT0: 247371 # score # COUNT1: write dead timebase program # COUNT2: lps # name of test item # elapsed: 30.003119 pid: 16803 # process ID# status: whether 0 # exited successfully # Pass "COUNT0: 2429" COUNT1: "COUNT2: lps# elapsed: 30.00289" pid: 503 "status: 0"# Pass" COUNT0: 24398 "COUNT1:" COUNT2: lps# elapsed: 30.00273 "pid: 21228 # status: 0*Dump score: 242919.0Count score: 243989.0Count score: 247371.0 > Results of 1 copy > score: 8188.34084738901 > > time: 30.0029255 > iterations: 2

The COUNT0,COUNT1,COUNT2 is calculated in this way, and the results of each process are as follows:

COUNT | x | y | x

After parsing, it becomes as follows:

COUNT0 = x COUNT1 = y COUNT2 = z

The scoring process is as follows:

According to the COUNT0 from small to big, get rid of the worst result of 1Accord 3.

If timebase, the time-consuming is all greater than 0, $product + = log ($COUNT0)-log (time-consuming / $timebase) otherwise you don't need to consider the time-consuming block: $product + = log ($COUNT0)

Individual performance score: $score = exp ($product/2), 2 is the statistical number of iterations, which is calculated to be exactly the same as the system output.

Its general algorithm idea: first through log dimensionality reduction, different times of different results as far as possible flat, average and then in the factorial, the following results are the same idea, first dimensionality reduction, then average, and then factorial.

Multi-process score

What if you execute multiple concurrently? Examples are as follows:

Shell Scripts (1 concurrent)-- 4 copies== > "/ opt/unixbench/UnixBench/pgms/looper" 60 "/ opt/unixbench/UnixBench/pgms/multi.sh" 12 > & 1 > > "/ opt/unixbench/UnixBench/results/VM_0_13_centos-2018-06-25-05.log" # Pass passport COUNT0: 461 cycles COUNT1: 6 miles COUNT2: lpm# elapsed: 60.00563 miles pid: 2085 miles status: stories COUNT0: 459 COUNT1: 6 months COUNT2: lpm# elapsed: 60. 009496# pid: 20859# status: 0# COUNT0: 4592# COUNT1: 60# COUNT2: lpm# elapsed: 60.011761# pid: 20862# status: 0# COUNT0: 4614# COUNT1: 60# COUNT2: lpm# elapsed: 60.011930# pid: 20864# status: 0#### Pass 2# COUNT0: 4547# COUNT1: 60# COUNT2: lpm# elapsed: 60.005597# pid: 10791# status: 0# COUNT0: 4590# COUNT1: 60# COUNT2: lpm# elapsed: 60.013270# pid: 10793# status: 0# COUNT0: 4578# COUNT1: 60# COUNT2: lpm# elapsed: 60.006054# pid: 10794# status: 0# COUNT0: 4561# COUNT1: 60# COUNT2: lpm# elapsed: 60.014214# pid: 10797# status: 0#### Pass 3# COUNT0: 4631# COUNT1: 60# COUNT2: lpm# elapsed: 60.013816# pid: 31734# status: 0# COUNT0: 4632# COUNT1: 60# COUNT2: lpm# elapsed: 60.012614# pid: 31735# status: 0# COUNT0: 4637# COUNT1: 60# COUNT2: Lpm# elapsed: 60.00563 "pid: 3173" status: "COUNT0: 464" COUNT1: 6 "COUNT2: lpm# elapsed: 60.00608" status: 0*Dump score: 18276.0Count score: 18416.0Count score: 18545.0 > Sum of 4 copies > score: 18477.4244713467 > time: 60.009621375 > iterations: 2

The scoring process is as follows:

Score: such as 18545, which is the sum of PASS3 items: 4631463246374645

Time spent: average, i.e. (60.013816060.012614060.0056333360.006082) / 40.060.00953625

Timebase: it's the same COUNT1

The above three steps are to synthesize a concurrent result into a result, and then integrate it in the same way as a single process.

Total score

An example of the legend score of multiple processes at the beginning of the article:

Each score calculated has a weighted score, which should be divided by 10 (as an individual competition, I think it doesn't matter whether it is divided by 10 or not, it is estimated to be close to the total score)

Total score calculation: = exp (average (each score score * 10)), and the result is exactly what we want.

In summary, the calculation of the score of UnixBench is comprehensively introduced, and the calculation process of excel proof is attached.

Attachment download: the process of calculating points. Xls. [Milazi] .1543830848.zip

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.