An example of performance with async_sinatra with thin and rainbows.

class AsyncTest < Sinatra::Base
  register Sinatra::Async

  enable :show_exceptions


  get '/test_sync' do
    "prova"
  end

  aget '/test_async' do 
    body { ["prova async"] }
  end
  
end

with rainbow and 4 worker_processes i obtain:

Requests per second:    805.41 [#/sec] (mean)
Time per request:       124.16 [ms] (mean)
Time per request:       1.24 [ms] (mean, across all concurrent requests)
Transfer rate:          95.10 [Kbytes/sec] received

with thin and one worker i obtain;

Requests per second:    1747.95 [#/sec] (mean)
Time per request:       57.21 [ms] (mean)
Time per request:       0.57 [ms] (mean, across all concurrent requests)
Transfer rate:          183.53 [Kbytes/sec] received

How can be possible this? What difference have thin and rainbows with eventmachine? I have also test
rainbows with one worker but this with only one worker is really fast.



RAINBOW

worker_processes 4

Rainbows! do
    use :EventMachine
    worker_connections 400
    client_max_body_size nil
    keepalive_timeout 0
end




ab -n 10000 -c 100 http://localhost:8080/test_async
This is ApacheBench, Version 1.3d <$Revision: 1.73 $> apache-1.3
Copyright (c) 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Copyright (c) 2006 The Apache Software Foundation, http://www.apache.org/

Benchmarking localhost (be patient)
Completed 1000 requests
Completed 2000 requests
Completed 3000 requests
Completed 4000 requests
Completed 5000 requests
Completed 6000 requests
Completed 7000 requests
Completed 8000 requests
Completed 9000 requests
Finished 10000 requests
Server Software:                                                           
Server Hostname:        localhost
Server Port:            8080

Document Path:          /test_async
Document Length:        11 bytes

Concurrency Level:      100
Time taken for tests:   12.416 seconds
Complete requests:      10000
Failed requests:        0
Broken pipe errors:     0
Total transferred:      1180708 bytes
HTML transferred:       110066 bytes
Requests per second:    805.41 [#/sec] (mean)
Time per request:       124.16 [ms] (mean)
Time per request:       1.24 [ms] (mean, across all concurrent requests)
Transfer rate:          95.10 [Kbytes/sec] received

Connnection Times (ms)
              min  mean[+/-sd] median   max
Connect:        0     1    2.8      0    66
Processing:     2   122   40.3    115   367
Waiting:        0   122   40.3    115   367
Total:          2   123   40.0    116   367

Percentage of the requests served within a certain time (ms)
  50%    116
  66%    130
  75%    141
  80%    148
  90%    172
  95%    199
  98%    241
  99%    263
 100%    367 (last request)
 
 
 
 THIN
 
 ab -n 10000 -c 100 http://localhost:3000/test_async
 This is ApacheBench, Version 1.3d <$Revision: 1.73 $> apache-1.3
 Copyright (c) 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
 Copyright (c) 2006 The Apache Software Foundation, http://www.apache.org/

 Benchmarking localhost (be patient)
 Completed 1000 requests
 Completed 2000 requests
 Completed 3000 requests
 Completed 4000 requests
 Completed 5000 requests
 Completed 6000 requests
 Completed 7000 requests
 Completed 8000 requests
 Completed 9000 requests
 Finished 10000 requests
 Server Software:        thin                                               
 Server Hostname:        localhost
 Server Port:            3000

 Document Path:          /test_async
 Document Length:        11 bytes

 Concurrency Level:      100
 Time taken for tests:   5.721 seconds
 Complete requests:      10000
 Failed requests:        0
 Broken pipe errors:     0
 Total transferred:      1050000 bytes
 HTML transferred:       110000 bytes
 Requests per second:    1747.95 [#/sec] (mean)
 Time per request:       57.21 [ms] (mean)
 Time per request:       0.57 [ms] (mean, across all concurrent requests)
 Transfer rate:          183.53 [Kbytes/sec] received

 Connnection Times (ms)
               min  mean[+/-sd] median   max
 Connect:        0     1    0.9      0    14
 Processing:    22    56   31.2     45   211
 Waiting:       17    55   31.3     45   211
 Total:         22    56   31.3     46   211

 Percentage of the requests served within a certain time (ms)
   50%     46
   66%     48
   75%     51
   80%     57
   90%     74
   95%    166
   98%    175
   99%    187
  100%    211 (last request)