Walid A. Hanafy
YOU?
Author Swipe
View article: LLM-Driven Auto Configuration for Transient IoT Device Collaboration
LLM-Driven Auto Configuration for Transient IoT Device Collaboration Open
Today's Internet of Things (IoT) has evolved from simple sensing and actuation devices to those with embedded processing and intelligent services, enabling rich collaborations between users and their devices. However, enabling such collabo…
View article: Data Centers Carbon Emissions at Crossroads: An Empirical Study
Data Centers Carbon Emissions at Crossroads: An Empirical Study Open
Data centers' energy consumption is expected to rise significantly over the next five years due to the accelerated growth of AI and its computational demands. Simultaneously, initiatives are underway to mitigate the environmental impact of…
View article: MEL: Multi-level Ensemble Learning for Resource-Constrained Environments
MEL: Multi-level Ensemble Learning for Resource-Constrained Environments Open
AI inference at the edge is becoming increasingly common for low-latency services. However, edge environments are power- and resource-constrained, and susceptible to failures. Conventional failure resilience approaches, such as cloud failo…
View article: FailLite: Failure-Resilient Model Serving for Resource-Constrained Edge Environments
FailLite: Failure-Resilient Model Serving for Resource-Constrained Edge Environments Open
Model serving systems have become popular for deploying deep learning models for various latency-sensitive inference tasks. While traditional replication-based methods have been used for failure-resilient model serving in the cloud, such m…
View article: GreenThrift: Optimizing Carbon and Cost for Flexible Residential Loads
GreenThrift: Optimizing Carbon and Cost for Flexible Residential Loads Open
Reducing buildings’ carbon emissions is an important sustainability challenge. While scheduling flexible building loads has been previously used for a variety of grid and energy optimizations, carbon footprint reduction using such flexible…
View article: CarbonEdge: Leveraging Mesoscale Spatial Carbon-Intensity Variations for Low Carbon Edge Computing
CarbonEdge: Leveraging Mesoscale Spatial Carbon-Intensity Variations for Low Carbon Edge Computing Open
The proliferation of latency-critical and compute-intensive edge applications is driving increases in computing demand and carbon emissions at the edge. To better understand carbon emissions at the edge, we analyze granular carbon intensit…
View article: CDN-Shifter: Leveraging Spatial Workload Shifting to Decarbonize Content Delivery Networks
CDN-Shifter: Leveraging Spatial Workload Shifting to Decarbonize Content Delivery Networks Open
View article: LACS: Learning-Augmented Algorithms for Carbon-Aware Resource Scaling with Uncertain Demand
LACS: Learning-Augmented Algorithms for Carbon-Aware Resource Scaling with Uncertain Demand Open
Motivated by an imperative to reduce the carbon emissions of cloud data centers, this paper studies the online carbon-aware resource scaling problem with unknown job lengths (OCSU) and applies it to carbon-aware resource scaling for execut…
View article: Going Green for Less Green: Optimizing the Cost of Reducing Cloud Carbon Emissions
Going Green for Less Green: Optimizing the Cost of Reducing Cloud Carbon Emissions Open
The continued exponential growth of cloud datacenter capacity has increased awareness of the carbon emissions when executing large compute-intensive workloads. To reduce carbon emissions, cloud users often temporally shift their batch work…
View article: CarbonScaler: Leveraging Cloud Workload Elasticity for Optimizing Carbon-Efficiency
CarbonScaler: Leveraging Cloud Workload Elasticity for Optimizing Carbon-Efficiency Open
Cloud platforms are increasing their emphasis on sustainability and reducing their operational carbon footprint. A common approach for reducing carbon emissions is to exploit the temporal flexibility inherent to many cloud workloads by exe…
View article: The War of the Efficiencies: Understanding the Tension between Carbon and Energy Optimization
The War of the Efficiencies: Understanding the Tension between Carbon and Energy Optimization Open
Major innovations in computing have been driven by scaling up computing infrastructure, while aggressively optimizing operating costs. The result is a network of worldwide datacenters that consume a large amount of energy, mostly in an ene…
View article: Understanding the Benefits of Hardware-Accelerated Communication in Model-Serving Applications
Understanding the Benefits of Hardware-Accelerated Communication in Model-Serving Applications Open
It is commonly assumed that the end-to-end networking performance of edge offloading is purely dictated by that of the network connectivity between end devices and edge computing facilities, where ongoing innovation in 5G/6G networking can…
View article: Ecovisor: A Virtual Energy System for Carbon-Efficient Applications
Ecovisor: A Virtual Energy System for Carbon-Efficient Applications Open
Cloud platforms' rapid growth is raising significant concerns about their carbon emissions. To reduce emissions, future cloud platforms will need to increase their reliance on renewable energy sources, such as solar and wind, which have ze…
View article: Model-driven Cluster Resource Management for AI Workloads in Edge Clouds
Model-driven Cluster Resource Management for AI Workloads in Edge Clouds Open
Since emerging edge applications such as Internet of Things (IoT) analytics and augmented reality have tight latency constraints, hardware AI accelerators have been recently proposed to speed up deep neural network (DNN) inference run by t…
View article: Table of contents
Table of contents Open
View article: A New Infrastructure Elasticity Control Algorithm for Containerized Cloud
A New Infrastructure Elasticity Control Algorithm for Containerized Cloud Open
In the last decade, containers have become a superior alternative to hypervisor-based virtualization. Containerization has revolutionized data centers from being an infrastructure-oriented to be application oriented. Modern cloud consumpti…