Interference-aware cloud performance management
(June 2015 - present)
Cloud computing provides economical and elastic resources to users for their software deployment needs; the low resource cost is realized via multi-tenancy, which allows for physical resources to be shared among several users via virtualization. Unfortunately, multi-tenancy results in undesirable performance effects, the most severe of which is interference, which is caused by lack
of effective performance isolation among users of colocated resources. Interference can result in significant performance degradation, increasing application response times by factors
The goal of this project is to enable user-centric solutions to detect and estimate interference in cloud-deployed applications without requiring any assistance from the hypervisor. These solutions will allow users to meet application performance targets in unreliable cloud environments.
- Improving server utilization via resource-adaptive batch VMs.
Seyyed Ahmad Javadi, Piyush Shyam Banginwar, Vaishali Chanana, Rashmi Narvekar, Mitesh Kumar Savita, Anshul Gandhi
Poster @ Middleware 2017 [pdf]
- Modeling and Analysis of Performance under Interference in the Cloud.
Scott Votke, Seyyed Ahmad Javadi, Anshul Gandhi
MASCOTS 2017 [pdf]
- DIAL: Reducing Tail Latencies for Cloud Applications via Dynamic Interference-aware Load Balancing.
Seyyed Ahmad Javadi, Anshul Gandhi
ICAC 2017 [pdf]
- Dynamic Interference-Aware Load Balancing.
Seyyed Ahmad Javadi, Himanshu Rajput, Anshul Gandhi
Poster @ SOCC 2016
- UIE: User-centric Interference Estimation for Cloud Applications
Seyyed Ahmad Javadi, Sagar Mehra, Bharath Kumar Reddy Vangoor, Anshul Gandhi
IC2E 2016 [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