Autoscaling cloud-deployed applications
(August 2014 - present)
Applications with a dynamic workload demand need access to a flexible infrastructure to meet performance guarantees and minimize resource costs. While cloud computing provides the elasticity
to scale the infrastructure on demand, cloud service providers lack control and visibility of user space applications, making it difficult to accurately scale the infrastructure. Thus, the burden of scaling falls on the user. That is, the user must determine when to trigger scaling
and how much to scale.
The goal of this project is to develop autoscaling solutions for cloud-deployed applications. Such solutions will lower resource rental costs for the users without compromising on performance. From the cloud service provider's perspective, such solutions will improve utilization and energy efficiency.
- Using Machine Learning for Black-Box Autoscaling.
Muhammad Wajahat, Alexei Karve, Andrzej Kochut, Anshul Gandhi
IGSC 2016 [pdf]
- Autoscaling for Hadoop Clusters
Anshul Gandhi, Parijat Dube, Andrzej Kochut, Li Zhang, Sidhartha Thota
IC2E 2016 [pdf] (also, Poster @ SOCC 2015 and Poster @ ICAC 2015 [pdf])
- The Unobservability Problem in Clouds
Anshul Gandhi, Parijat Dube, Alexei Karve, Andrzej Kochut, Harsha Ellanti
ICCAC 2015 [pdf]
© Copyright 2014-2017 PACE Lab, Stony Brook University