Dear John, More specifically, I'd been working on RA in distributed systems awhile, and also, I've presented a prototype for it. The most important thing that I've learnt is: talking about RA without regarding to workloads (e.g., HTC, MTC, BoT, Scientific, Data-Intensive, etc.) is not optimal at all.
So, I am thinking about providing different RA modules for each workload. It means, when the api gets a client's workload, first, some workload characterizations are needed (e.g., statistical modeling, classification, multi-queue, etc.), then, based on nature of the workload the RA process by using the appropriate RA module can be near optimal, I hope. By optimal, I mean less resource consumption as compared to a general RA for all workloads. Nevertheless, since I am coming from theory (i.e., simulation of RA for distributed systems) to here and I don't have a big image about RA in CloudStack in mind, I'm wondering if someone gives some clues to make me ready for putting my idea into practice. Also, unfortunately, I cannot attend in CloudStack Collab 2013, but I will follow its news and your slides. Best regards, Pouya ________________________________ From: John Burwell <jburw...@basho.com> To: dev@cloudstack.apache.org; Linux TUX <azgil.i...@yahoo.com> Sent: Friday, 21 June 2013, 17:26 Subject: Re: Resource Management/Allocation for CS Pouya, What problem/issue/enhancements do you have in mind? If you are attending CloudStack Collab 2013, I will speaking on this topic on Monday @ 2:30 (How to Run from a Zombie: CloudStack Distributed Process Management). My slides will be available online after the talk as well. Thanks, -John On Jun 21, 2013, at 8:47 AM, Linux TUX <azgil.i...@yahoo.com> wrote: > Hi All, > > Does anybody can tell me which parts of CloudStack's source code are related > to its Resource Allocation (RA) process? > By RA, I mean the part of code that is responsible for VM migration or > process migration, if there is any. > As you know, there are two kinds of RA, to wit: 1) server side such as VM > migration, or 2) client side such as clients' proprietary schedulers. > Furthermore, client side's RA's success is dependent on knowing sever side's > RA very well. > > So, since i am interested to work on RA of CloudStack, if, with regard to the > above information, you have any idea, please tell me? > Also, if your are working on it, please let me know. Finally, it would be > really approciated if you tell me which parts of the source code > is related to implementation of CloudStack's RA algorithms. > > Best regards, > Pouya