Vincent Lannurien
d822cdaa6a
All checks were successful
Build and deploy static pages / build-and-push (push) Successful in 19s
109 lines
3.6 KiB
Markdown
109 lines
3.6 KiB
Markdown
---
|
|
title: SEC 2024 Digest
|
|
venue: SEC
|
|
year: 2024
|
|
date: '2024-01-01'
|
|
tags: []
|
|
paper_count: 12
|
|
draft: false
|
|
---
|
|
|
|
12 papers selected.
|
|
|
|
---
|
|
|
|
### EdgeCore: Resource Dependency-Aware Multi-Tenant Orchestration for Mobile Edge Clouds
|
|
|
|
*Amran Haroon*
|
|
|
|
**TL;DR** — Introduces a multi-tenant edge orchestration system that captures resource dependencies across co-located workloads, demonstrating significant improvements in task completion latency and resource utilization.
|
|
|
|
---
|
|
|
|
### Righteous: Automatic Right-Sizing for Complex Edge Deployments
|
|
|
|
*Aniruddha Rakshit*
|
|
|
|
**TL;DR** — Presents an automated right-sizing framework for edge deployments that dynamically adjusts resource allocations to match workload demands without manual intervention.
|
|
|
|
---
|
|
|
|
### Colibri: Efficient Collection of Fine-Grained Resource Metrics Necessary for Mobile Edge Computing
|
|
|
|
*Ke-Jou Hsu*
|
|
|
|
**TL;DR** — Proposes a low-overhead monitoring system for collecting fine-grained resource metrics at the edge, enabling more accurate profiling for MEC scheduling decisions.
|
|
|
|
---
|
|
|
|
### HyperDrive: Scheduling Serverless Functions in the Edge-Cloud-Space 3D Continuum
|
|
|
|
*Thomas W. Pusztai*
|
|
|
|
**TL;DR** — Extends serverless scheduling across a three-dimensional edge-cloud-space continuum, addressing latency and resource constraints introduced by satellite and terrestrial tiers.
|
|
|
|
---
|
|
|
|
### Falcon: Live Reconfiguration for Stateful Stream Processing on the Edge
|
|
|
|
*Pritish Mishra*
|
|
|
|
**TL;DR** — Enables live, low-disruption reconfiguration of stateful stream processing pipelines at the edge, minimizing downtime during topology changes.
|
|
|
|
---
|
|
|
|
### FusedInf: Efficient Swapping of DNN Models for On-Demand Serverless Inference Services on the Edge
|
|
|
|
*Sifat Ut Taki*
|
|
|
|
**TL;DR** — Reduces cold-start latency for serverless DNN inference at the edge by fusing model loading with active inference through selective layer swapping.
|
|
|
|
---
|
|
|
|
### EcoEdgeInfer: Dynamically Optimizing Latency and Sustainability for Inference on Edge Devices
|
|
|
|
*Sri Pramodh Rachuri*
|
|
|
|
**TL;DR** — Co-optimizes inference latency and energy sustainability on edge devices by dynamically trading off accuracy and hardware utilization under carbon-aware constraints.
|
|
|
|
---
|
|
|
|
### Elastic Execution of Multi-Tenant DNNs on Heterogeneous Edge MPSoCs
|
|
|
|
*Soroush Heidari*
|
|
|
|
**TL;DR** — Demonstrates elastic, interference-aware co-execution of multiple DNNs across heterogeneous processing elements in edge MPSoCs to maximize throughput and fairness.
|
|
|
|
---
|
|
|
|
### Optimizing Edge Offloading Decisions for Object Detection
|
|
|
|
*Jiaming Qiu*
|
|
|
|
**TL;DR** — Formulates and solves an online offloading decision problem for object detection that jointly minimizes latency and energy consumption under variable network conditions.
|
|
|
|
---
|
|
|
|
### VideoJam: Self-Balancing Architecture for Live Video Analytics
|
|
|
|
*Youssouph Faye*
|
|
|
|
**TL;DR** — Proposes a self-balancing edge architecture for live video analytics that dynamically redistributes pipeline stages to prevent bottlenecks under fluctuating camera workloads.
|
|
|
|
---
|
|
|
|
### OVIDA: Orchestrator for Video Analytics on Disaggregated Architecture
|
|
|
|
*Manavjeet Singh*
|
|
|
|
**TL;DR** — Designs an orchestration layer for disaggregated edge hardware that places and migrates video analytics microservices to exploit spatial locality and heterogeneous accelerators.
|
|
|
|
---
|
|
|
|
### TA-ASF: Attention-Sensitive Token Sampling and Fusing for Visual Transformer Models on the Edge
|
|
|
|
*Junquan Chen*
|
|
|
|
**TL;DR** — Accelerates Vision Transformer inference at the edge by pruning and fusing attention tokens based on saliency, achieving accuracy-efficiency trade-offs suitable for resource-constrained devices.
|
|
|