I am the Computing & Climate Impact Fellow at MIT Climate & Sustainability Consortium (MCSC) and a postdoc with Christina Delimitrou at MIT Computer Science & Artificial Intelligence Laboratory (CSAIL).
I work on improving the sustainability of computing and AI. I demonstrate that the status quo approach to computing sustainability, i.e., the lone focus on improving energy efficiency, is insufficient to tackle the emerging dynamics of saturating efficiency gains, exponential growth in computing demand, and electricity availability constraints. To solve this problem, I design systems, algorithms, and metrics that elevate sustainability to a first-class objective in computer system design and operation.
I have made fundamental contributions to the design of sustainable computer systems through multiple artifacts, such as EcoVisor, CarbonContainers, CarbonScaler, and WattScope. One of my solutions to improving resource efficiency in datacenters is deployed on all the Google datacenters serving applications such as Google Search, Gmail, and YouTube.
Before joining MIT, I was at UMass Amherst for 7+ years working with Prashant Shenoy as a postdoc and completing my PhD in Computer Engineering (2022) at the Sustainable Computing Lab with David Irwin. During my PhD, I worked as a Student Researcher on Google’s Borg Scheduler mentored by Nan Deng and Krzysztof Rządca. I also worked at VMWare Research Group on Computing Sustainability Benchmarking.
Before joining UMass, I worked as a Research Assistant at the Advanced Communication Lab at LUMS, Pakistan, and Systems & Networks Lab at NUCES-NU, Pakistan. I did my MS in Energy Systems Engineering from Center for Advanced Studies in Energy at NUST, Pakistan and BS in Electrical (Power) Engineering at UET Lahore, Pakistan.
I take a huge pride academic service and was recognized as one of the top three reviewers at ACM e-Energy’23.
Job Market: I am currently seeking tenure-track faculty positions and am excited to share my work with potential colleagues and collaborators. You can find my Research Statement and CV for a closer look at my background and contributions. If you’re hiring or interested in learning more about my work, let’s talk!
Selected Publications
- SoCC 2024"The Sunk Carbon Fallacy: Rethinking Carbon Footprint Metrics for Effective Carbon-Aware Scheduling" -- ACM Symposium on Cloud Computing (SoCC) , 2024. PDF
- EuroSys 2024"On the Limitations of Carbon-Aware Temporal and Spatial Workload Shifting in the Cloud" -- European Conference on Computer Systems (EuroSys) , 2024. PDF
- ASPLOS 2024"Going Green for Less Green: Optimizing the Cost of Reducing Cloud Carbon Emissions" -- ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) , 2024. PDF
- ICML 2024"Chasing Convex Functions with Long-term Constraints" -- The International Conference on Machine Learning (ICML) , 2024. PDF
- e-Energy 2023"Jointly Managing Electrical and Thermal Energy in Solar- and Battery-powered Computer Systems" -- ACM International Conference on Future and Sustainable Energy Systems (e-Energy) , 2023. PDF
- EuroSys 2021"Take it to the Limit: Peak Prediction-driven Resource Overcommitment in datacenters" -- European Conference on Computer Systems (EuroSys) , 2021. PDF
Latest News
09/25/2024: My paper titled The Sunk Carbon Fallacy: Rethinking Carbon Footprint Metrics for Effective Carbon-Aware Scheduling has been accepted to ACM SoCC 2024. |
09/08/2024: I, with Prof. Elsa Olivetti, received a Nokia research grant of $57,000 for our research on Provisioning and Operating Sustainable Datacenters. |
07/25/2024: I will be talking about the climate and sustainability implications of generative AI at 2024 MIT ILP Sustainability Conference on October 22, 2024. |
06/25/2024: Our paper CarbonScaler: Leveraging Cloud Workload Elasticity for Optimizing Carbon-Efficiency won the Best Student Paper Award at ACM SIGMETRICS 2024. |
06/07/2024: Our paper The Green Mirage: Impact of Location- and Market-based Carbon Intensity Estimation on Carbon Optimization Efficacy was Best Full Paper Runner Up at ACM e-Energy 2024. |