4 papers
MQFQ-Sticky: Fair Queueing For Serverless GPU Functions
Alexander Fuerst, Siddharth Anil, Vishakha Dixit +3
Hardware accelerators like GPUs are now ubiquitous in data centers, but are not fully supported by common cloud abstractions such as Functions as a Service (FaaS). Many popular and…
OmniLearn: A Framework for Distributed Deep Learning over Heterogeneous Clusters
Sahil Tyagi, Prateek Sharma
Deep learning systems are optimized for clusters with homogeneous resources. However, heterogeneity is prevalent in computing infrastructure across edge, cloud and HPC. When traini…
The Jevons Paradox In Cloud Computing: A Thermodynamics Perspective
Prateek Sharma
How do we explain the simultaneous growth in energy efficiency of cloud computing and its energy consumption? The Jevons paradox provides one perspective of this phenomenon. Howeve…
FaasMeter: Energy-First Serverless Computing
Abdul Rehman, Alexander Fuerst, Prateek Sharma
Functions as a Service has emerged as a popular abstraction for a wide range of cloud applications and an important cloud workload. We present the design and implementation of Faas…