Cloud computing allows tenants to economically rent compute and storage resources from providers. To enable low resource prices, providers consolidate multiple tenants onto a single physical server. However, this sharing of physical resources among tenants often leads to contention, resulting in unpredictable performance. Worse, tenants cannot observe resource contention due to the opaque nature of cloud computing. This project will develop novel performance models to estimate resource contention in opaque cloud deployments. These models will then be leveraged to develop solutions for cloud tenants that mitigate performance variation, thus enabling predictable performance in clouds.
Collaborators:
Publications:
- Leveraging Queueing Theory and OS Profiling to Reduce Application Latency
Anshul Gandhi, Amoghavarsha Suresh
Tutorial (invited) @ Middleware 2019 [pdf, slides]
- Scavenger: A black-box batch workload resource manager for improving utilization in cloud environments
Seyyed Ahmad Javadi, Muhammad Wajahat, Amoghavarsha Suresh, Anshul Gandhi
SOCC 2019 [pdf]
- Optimal Markovian Dynamic Control of Interference-Prone Server Farms
Scott Votke, Jazeem Abdul Jaleel, Amoghavarsha Suresh, Mohammad Delasay, Sherwin Doroudi, Anshul Gandhi
MASCOTS 2019 [pdf]
- Tighter Lyapunov Truncation for Multi-Dimensional Continuous Time Markov Chains with Known Moments
Gagan Somashekar, Mohammad Delasay, Anshul Gandhi
MAMA 2019 [pdf]
- Using Variability as a Guiding Principle to Reduce Latency in Web Applications via OS Profiling.
Amoghavarsha Suresh, Anshul Gandhi
WWW 2019 [pdf]
- 5GCoreLite: Scalable and Resource Efficient Next Generation Cellular Packet Core.
Vasudevan Nagendra, Arani Bhattacharya, Anshul Gandhi, Samir Das
Poster @ NSDI 2019 [pdf]
- Scalable and Resource Efficient Control Plane for Next Generation Cellular Packet Core.
Vasudevan Nagendra, Arani Bhattacharya, Anshul Gandhi, Samir Das
SOSR 2019 [pdf]
Students:
Sponsors:
