Baunsgaard opened a new pull request, #1926:
URL: https://github.com/apache/systemds/pull/1926
This PR contains a bit of a controversy.
At startup the first thing we do is to call Hadoop to parse the Hadoop
specific arguments.
This takes ~ 200 ms at startup before we start our timing of SystemDS.
before:
```txt
1,6187 +- 0,0766 seconds time elapsed ( +- 4,73% )
```
after:
```txt
1,4366 +- 0,0496 seconds time elapsed ( +- 3,46% )
```
The downside of this modification is that Hadoop arguments are not parsed to
the Hadoop file system, that said, i am unaware if the arguments actually work
at the moment because ---to the best of my knowledge---we have not tested if
the system work when there are more than one HDFS system available.
The script used for the test is :
```R
print("Hello World!")
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
