Hi Tuna, You can find description of sample autoscaling policy from [1]
[1] https://cwiki.apache.org/confluence/display/STRATOS/4.1.0+Sample+Auto-scaling+Policy+Definition Please find some of the answers inline. On Sun, Dec 28, 2014 at 8:58 PM, Nguyen Anh Tu <[email protected]> wrote: > Dear guys, > > Could anyone share detail of the current deployment of autoscaling > feature? I found some documents on cwiki but they are too general. > > Some questions: > - what are "gradient" & "second derivative"? > We have calculate gradient and second derivatives of memory consumption and the load average of the instances. After instances get activated it start to published the health stats. Based on the stats CEP calculates the gradient and second derivative based on the stats received. CEP waits for one minute time interval and calculates those values. > - when scaling down, do we terminate instances immediately? > No. Autoscaler will wait for count down. After that it terminate the instance. > - what is difference between autoscale policy and deployment policy? > - do we have "scheduler time" for autoscaling? > > Thanks for help, > --Tuna > -- Regards, Manula Chathurika Thantriwatte Software Engineer WSO2 Inc. : http://wso2.com lean . enterprise . middleware email : [email protected] / [email protected] phone : +94 772492511 blog : http://manulachathurika.blogspot.com/
