Files
SearchServices/e2e-test/qa/benchmark/README.md
T
2019-03-20 15:19:19 +00:00

42 lines
1.6 KiB
Markdown

## About
Starting automatically Benchmark infrastructure:
* [BM Manager](https://github.com/Alfresco/alfresco-bm-manager)
* [BM Drivers](https://github.com/Alfresco/alfresco-bm-manager/tree/master/docs/bm-driver)
* running create [users](https://github.com/Alfresco/alfresco-bm-load-users)
* running create [data](https://github.com/Alfresco/alfresco-bm-load-data)
* running [rest test](https://github.com/Alfresco/alfresco-bm-rest-api)
## Prerequisites
* ACS deployed in AWS (EKS or Docker)
* Benchmark knowledge (check the official documentation above)
## How to use it?
### a) start BM Manager and Drivers locally
```shell
$ make start-all
```
>check the official documentation how to use the benchmark
### b) prepare data for benchmarking
> we need to create users/sites and load some content in order to tests the search performance
```shell
$ make ALFRESCO_URL=<http://alserver:port> benchmark-prepare
```
>this will call some shell scripts
### c) run the actual benchmark tests
>you can tweak the benchmark tests as you want in the shell script executed by this task
```shell
$ make ALFRESCO_URL=<http://alserver:port> benchmark-run
```
### `!` Bamboo Build
We defined a sad benchmark build: [SAD-BEN](https://bamboo.alfresco.com/bamboo/browse/SAD-BEN) that will execute the steps above automatically for you.
If you run it customized, you can override the `ALFRESCO_URL` variable to point to a deployed ACS instance (default to: "http://ec2-34-245-148-19.eu-west-1.compute.amazonaws.com:8081")
At the end of the tests you will have access to Excel reports (the actual benchmark results)