On Fri, Dec 6, 2013 at 12:33 PM, Lahiru Sandaruwan <[email protected]> wrote:
> > > > On Fri, Dec 6, 2013 at 12:20 PM, Nirmal Fernando > <[email protected]>wrote: > >> Ah.. yes.. all the data Autosclaer receives.. for an example, based on >> the health stats you receive from members, we can build up an accumulated >> state of all partitions and then if we need to spawn a new instance of a >> cluster, we could select the less loaded partition. >> > > Can you explain why we should consider health stat details on selecting > the next partition? > To make partition selection intelligent. > > > Do you mean the number of instances in a particular partition by "load of > the partition"? > > Or the load of the instances of a particular partition? > As I said, I meant the accumulated health of the partition. May be CEP could send partition wise aggregated health stats. > > > >> >> >> On Fri, Dec 6, 2013 at 12:10 PM, Lahiru Sandaruwan <[email protected]>wrote: >> >>> What do you mean by "data it receives"? >>> >>> >>> On Fri, Dec 6, 2013 at 12:07 PM, Nirmal Fernando <[email protected] >>> > wrote: >>> >>>> Didn't quite get your question. Can you please elaborate? >>>> >>>> >>>> On Fri, Dec 6, 2013 at 12:04 PM, Lahiru Sandaruwan <[email protected]>wrote: >>>> >>>>> Nirmal, >>>>> >>>>> Do you mean CPU/ Memory consumption details or the current >>>>> topology(The partitions of the current instances)? >>>>> >>>>> >>>>> On Fri, Dec 6, 2013 at 11:58 AM, Nirmal Fernando < >>>>> [email protected]> wrote: >>>>> >>>>>> All, >>>>>> >>>>>> Currently when a new member is about to spawn, the partition would >>>>>> always be selected via an algorithm and there's no much intelligence over >>>>>> there. Ideally, what should happen is; autoscaler should identify the >>>>>> partition that it needs to spawn the next instance via the data it >>>>>> receives, if it fails only it should pick next partition up from a >>>>>> algorithm. >>>>>> wdyt? >>>>>> >>>>>> -- >>>>>> Best Regards, >>>>>> Nirmal >>>>>> >>>>>> Nirmal Fernando. >>>>>> PPMC Member & Committer of Apache Stratos, >>>>>> Senior Software Engineer, WSO2 Inc. >>>>>> >>>>>> Blog: http://nirmalfdo.blogspot.com/ >>>>>> >>>>> >>>>> >>>>> >>>>> -- >>>>> -- >>>>> Lahiru Sandaruwan >>>>> Software Engineer, >>>>> Platform Technologies, >>>>> WSO2 Inc., http://wso2.com >>>>> lean.enterprise.middleware >>>>> >>>>> email: [email protected] cell: (+94) 773 325 954 >>>>> blog: http://lahiruwrites.blogspot.com/ >>>>> twitter: http://twitter.com/lahirus >>>>> linked-in: http://lk.linkedin.com/pub/lahiru-sandaruwan/16/153/146 >>>>> >>>>> >>>> >>>> >>>> -- >>>> Best Regards, >>>> Nirmal >>>> >>>> Nirmal Fernando. >>>> PPMC Member & Committer of Apache Stratos, >>>> Senior Software Engineer, WSO2 Inc. >>>> >>>> Blog: http://nirmalfdo.blogspot.com/ >>>> >>> >>> >>> >>> -- >>> -- >>> Lahiru Sandaruwan >>> Software Engineer, >>> Platform Technologies, >>> WSO2 Inc., http://wso2.com >>> lean.enterprise.middleware >>> >>> email: [email protected] cell: (+94) 773 325 954 >>> blog: http://lahiruwrites.blogspot.com/ >>> twitter: http://twitter.com/lahirus >>> linked-in: http://lk.linkedin.com/pub/lahiru-sandaruwan/16/153/146 >>> >>> >> >> >> -- >> Best Regards, >> Nirmal >> >> Nirmal Fernando. >> PPMC Member & Committer of Apache Stratos, >> Senior Software Engineer, WSO2 Inc. >> >> Blog: http://nirmalfdo.blogspot.com/ >> > > > > -- > -- > Lahiru Sandaruwan > Software Engineer, > Platform Technologies, > WSO2 Inc., http://wso2.com > lean.enterprise.middleware > > email: [email protected] cell: (+94) 773 325 954 > blog: http://lahiruwrites.blogspot.com/ > twitter: http://twitter.com/lahirus > linked-in: http://lk.linkedin.com/pub/lahiru-sandaruwan/16/153/146 > > -- Best Regards, Nirmal Nirmal Fernando. PPMC Member & Committer of Apache Stratos, Senior Software Engineer, WSO2 Inc. Blog: http://nirmalfdo.blogspot.com/
