Skip to content

Instantly share code, notes, and snippets.

@miguelbalparda
Created February 2, 2016 13:05
Show Gist options
  • Save miguelbalparda/bf6f2dea0e60c2504b4d to your computer and use it in GitHub Desktop.
Save miguelbalparda/bf6f2dea0e60c2504b4d to your computer and use it in GitHub Desktop.
** SIEGE 3.0.6
** Preparing 500 concurrent users for battle.
The server is now under siege...
Lifting the server siege...      done.
Transactions:                  39116 hits
Availability:                 100.00 %
Elapsed time:                  59.44 secs
Data transferred:             541.05 MB
Response time:                  0.26 secs
Transaction rate:             658.08 trans/sec
Throughput:                     9.10 MB/sec
Concurrency:                  168.92
Successful transactions:       39042
Failed transactions:               0
Longest transaction:            6.26
Shortest transaction:           0.07
@miguelbalparda
Copy link
Author

siege -c 500 -i -t60s urls.txt
urls.txt was generated using this post

@miguelbalparda
Copy link
Author

Type : Community
Version : 1.9.0.1
SKU Count : 3974

Running on MCE-SIP-200

@litespeedtech
Copy link

Thanks for the result.

Can you please try

siege -b -c 50 -n 10 -i -f urls.txt
siege -b -c 100 -n 10 -i -f urls.txt
siege -b -c 500 -n 10 -i -f urls.txt

Those command simulates 50, 100 and 500 visitors, each visitor view 10 pages.

What you did with siege is simulating 500 concurrent users, each visitor view 78 pages average, In a real-world situation, it is not likely each visitor has that many page views.

I assuming urls.txt only contain category and product detail pages, no static assets, static asset will boost TPS.

Please do the same test without varnish, see how much you can get. We can get a rough idea how much varnish improves the performance.

That's a nice test bed. pretty powerful.

Thanks,
George Wang

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment