| Paper | Keywords | Institute (first) | Publication | Others |
|---|---|---|---|---|
| A survey and measurement study of GPU DVFS on energy conservation | DVFS survey | HKBU | Digital Communications and Networks | |
| Characterizing Power Management Opportunities for LLMs in the Cloud | LLM | Microsoft Azure | ASPLOS 2024 | |
| Predict; Don’t React for Enabling Efficient Fine-Grain DVFS in GPUs | Fine Grain DVFS | MicroSoft | ASPLOS 2024 | |
| Roofline-aware DVFS for GPUs | roofline DVFS | Eindhoven University of Technology | ADAPT 2014 | |
| Effects of Dynamic Voltage and Frequency Scaling on a K20 GPU | Marquette University | ICPP 2013 | Key Points: 1. MM benchmark shows performance is linearly proportional to the GPU frequency when using a high memory speed 2. CPU DVFS impacts the CPU and other system components | |
| Evaluating the energy impact of device parameters for DNN inference on edge | DNN inference； edge device | Stony Brook University | IGSC 2023 | Key Points: default parameter setting is not energy optimal; Power consumption increases linearly with frequency. The decrease in inference latency starts to plateau out at higher frequency as frequency is not the bottkeneck and performance is limited by other components such as memory and I/O bandwidth |
| Zeus: Understanding and Optimizing GPU Energy Consumption of DNN Training | DNN training | University of Michigan | NSDI 2023 | |
| The Design, Implementation and Evaluation of a Compiler Algorithm for GPU energy Reduction | Programming language; CPU energy | Rutgers, The State University of New Jersey | PLDI 2003 | |
| A Dynamic Compilation Framework for Controlling Microprocessor Energy and Performance | Compiler | Princeton University | MICRO 2005 | |
| Energy-Aware Non-Preemptive Task Scheduling With Deadline Constraint in DVFS-Enabled Heterogeneous Clusters | Task scheduling; Deadline Constraint | HKBU | TPDS 2022 | |
| EnvPipe: Performance-preserving DNN Training Framework for Saving Energy | Pipeline parallelism; Training | KAIST | ATC 2023 | |
| Power-aware Deep Learning Model Serving with μ-Serve | LLM Serving | UIUC | ATC 2024 | |
| DynamoLLM: Designing LLM Inference Clusters for Performance and Energy Efficiency | LLM Serving Cluster | UIUC | Arxiv | |
| Improving GPU Energy Efficiency through an Application-transparent Frequency Scaling Policy with Performance Assurance | Energy Efficiency | PCL,HKBU | Eurosys 2024 | |
| GPGPU Power Estimation with Core and Memory Frequency Scaling | estimate power with frequency | HKBU | ACM SIGMETRICS Performance Evaluation Review | |
| GPGPU performance estimation with core | estimate performance with frequency | HKBU | TPDS 2020 |
-
Notifications
You must be signed in to change notification settings - Fork 0
galeselee/DVFS_PaperList
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
Energy is a very noticable topic. Dynaimc Voltage and Frequency Scaling is a technique for CPU and GPU power consumption. Here is a paperlist of DVFS and power consumption.
Topics
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published