Improving responsiveness of autoscaling systems

Gregory Nichols


Supervised by Omer F Rana; Moderated by Hantao Liu

The aim of this project would be to implement more efficient and responsive autoscaling for the existing production parsers for a company I have connections with. Through this project I hope to: (i) improve the deployment model for autoscaling in terms of reactivity, granularity and termination speed; (ii) implement a system to predict and anticipate increased workload and scale up accordingly. There would also be potential to extend this project further by extending the deployment model to the whole system in order to prevent bottlenecks other system areas.

Initial Plan (30/01/2017) [Zip Archive]

Final Report (05/05/2017) [Zip Archive]

Publication Form