Improvements to Autoscaling in Apache Stratos

You can find some details about autoscaling from [1] and [2].

1. Improve Autoscaling to predict the number of instances required in the next time interval.

Currently we predict the load for next time interval. Then we use a threshold to decide on scale up or down. Factors that we are considering are the requests in flight at load balancer and Load average and memory consumption of cartridge instance.

Best approach would be to decide the number of instances that will be required to handle the load in next time interval.

2. Predict the load according to a schedule defined by end user

This aspect will be needed to handle seasonal load expectations.

E.g. High load on Christmas and new year period.

[1] http://www.sc.ehu.es/ccwbayes/isg/administrator/components/com_jresearch/files/publications/autoscaling.pdf

[2] http://docs.aws.amazon.com/AutoScaling/latest/DeveloperGuide/WhatIsAutoScaling.html

Skills needed :

Java, Drools, SOA, Json, AMQP messaging protocol