find and solve the bottleneck of initializing netCDF files

[Reimar Bauer]

We have to rework the idea to initialize the get capabilities XML file. This needs to be done extremly fast compared to now.

Now sometimes we run in the defined 30 seconds timeout. I would prefer a solution where only scrolling through a list of servers will immediatly update the capabilities informations, without clicking an additional button.

[Reimar Bauer]

related #221

[Reimar Bauer]

by an apache benchmark test, about 100 requests, there is no bottleneck in reading netCDF files.

This is ApacheBench, Version 2.3 <$Revision: 1706008 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/
Benchmarking staging (be patient).....done
Server Software:        Apache/2.4.18
Server Hostname:        staging
Server Port:            80
Document Path:          /demo?service=WMS&request=GetCapabilities&version=1.1.1
Document Length:        15007 bytes
Concurrency Level:      1
Time taken for tests:   22.477 seconds
Complete requests:      100
Failed requests:        0
Total transferred:      1518400 bytes
HTML transferred:       1500700 bytes
Requests per second:    4.45 [#/sec] (mean)
Time per request:       224.773 [ms] (mean)
Time per request:       224.773 [ms] (mean, across all concurrent requests)
Transfer rate:          65.97 [Kbytes/sec] received
Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:        0    1   1.2      1      11
Processing:   205  224  24.8    220     425
Waiting:      203  223  24.8    219     424
Total:        205  225  25.7    221     436
Percentage of the requests served within a certain time (ms)
  50%    221
  66%    224
  75%    227
  80%    229
  90%    233
  95%    244
  98%    331
  99%    436
 100%    436 (longest request)

[Reimar Bauer]

reworking the UI is an other issue, but there is not a bottleneck by current code