Definition of the attribute _size in the container class

[Yung Lam]

I am confuse with the attribute _size in the container class from the container.py, as in line 63: self._size = data.pop('size', 1), so the _size of the container can be greater than 1, but beside the occupation of machine quota for container number, that stated in machine.py line 347:

    def availability(self):

        """ Free Machine Capacity. """

        return self._size - sum(c.size for c in self._containers)

will the attribute _size in the container affect other setting of Rapyuta? such as the CPU and memory resource allocation? Or priority of cpu job? thank

 

[Dominique Hunziker]

This attribute is not really used at the moment. The idea is that a container with a bigger size gets more resources (CPU, memory) allocated than a container with a smaller size. However, at the moment the argument is ignored by 'rce-container', which is responsible for launching the containers. Therefore, it is only used to limit the number of containers which can be run simultaneously on one machine. There are also some other parts especially in the Machine/Container class which are (partially) implemented, but are not used in the 'rce-container' process.

[Yung Lam]

So this feature is a #TODO？ 

By the way, do Rapyuta developer accept pull request in github for improvement of Rapyuta contributed by others?

Dominique Hunziker於 2014年1月17日星期五UTC+8上午1時29分52秒寫道：

[Dominique Hunziker]

Yes, this feature is not yet finished.
Yes, we accept pull requests and will merge improvements.