AB testing tool as record

Recently, a test was conducted on the new architecture of jujuya, with 200 requests and 20 concurrent requests. this test intensity accounts for 70-80% of the CPU, and w3p occupies more than 70 MB of memory. I want to test it several times to see if its memory will increase. I can't help without a test machine, developers need to work. I guess there is a problem with the CPU and performance needs to be optimized.
Mark the usage of the testing tool by the way

Basic usage :
AB-n Total requests -C Concurrency Test URL
Example : AB-N 1000-C 50 http://www.newdev.gztest.com /
Server software: Microsoft-Microsoft IIS/7.0
Server Hostname: Www.newdev.gztest.com
Server port: 80
Document path:
Document length: 82522 bytes # Request Document size

Concurrency level: 50 # Concurrency
Time taken for tests: 92.76140 seconds # Time consumed for completing all requests
Complete requests: 10000 # Total requests
Failed requests: 1974 # Failed requests
(CONNECT: 0, Length: 1974, exceptions: 0)
Write errors: 0
Total transferred: 827019400 bytes # Total Transfer size
HTML transferred: 825219400 bytes // The amount of HTML content transmitted throughout the scenario
Requests per secondd: 108.61 [#/sec] (mean )# Requests per second ( Average ) //One of the most important indicators is equivalentLRInTransactions per second, In bracketsMeanThis is an average value.
Time per request: 460.381 [MS] (mean )# Time of each concurrent request ( All concurrency ) //The second indicator that everyone cares most about is equivalentLRInAverage Transaction Response Time, In bracketsMeanThis is an average value.
Time per request: 9.208 [MS] (mean, internal SS all concurrent requests )# Time of each request ( Concurrency average ) // Average actual running time of each request
Transfer Rate: 8771.39 [Kbytes/sec] received # Transmission Rate //The average network traffic per second can help eliminate the problem of long response time due to excessive network traffic
Percentage of the requests served within a certain time (MS)
50% 2680
66% 2806
75% 2889
80% 2996
90% 11064
95% 20161
98% 21092
99% 21417
100% 21483 (longest request)
// Response of all requests in the entire scenario. In the scenario, each request has a response time, of which 50% of the user response time is less 2680 In milliseconds, 60% of user response times are less 2806 Millisecond, the maximum response time is less 21417 Millisecond For concurrent requests, the CPU does not actually process them at the same time, but is rotated one by one based on the time slice obtained by each request, therefore, the first time per request is basically equal to the second time per request time multiplied by the number of concurrent requests.

Connection times (MS )# Connection time
Min mean [+/-SD] median Max
Connect (# Connection ): 0 0 2.1 0 46
Processing (# Processing ): 31 458 94.7 438 1078
Waiting (# Wait ): 15 437 87.5 422 938
Total: 31 458 94.7 438

Other parameters:

-N requests Total requests
-C concurrency Concurrency
-T timelimit The time when the response is received.
-P postfile post Data Files
-T Content-Type post Content-Type
-V verbosity how much troubleshooting info to print
-W print out results in HTML tables
-I use head instead of get
-X attributes string to insert as table attributes
-Y attributes string to insert As tr attributes
-Z attributes string to insert as TD or th attributes
-C attribute Join Cookie, eg. 'apache = 1234. (repeatable)
-H attribute Join HTTP Header , Eg. 'Accept-encoding: gzip'
Inserted after all normal header lines. (repeatable)
-A attribute HTTP Verify , Separate user names and passwords
-P attribute add basic proxy authentication, the attributes
Are a colon separated username and password.
-X Proxy: Port Proxy Server
-V View AB Version
-K use HTTP keepalive feature
-d do not show percentiles served table.
-s do not show confidence estimators and warnings.
-G filename output collected data to gnuplot format file.
-e filename output CSV file with percentages served
-H display usage information (this message)