The following content is copyrighted from NSDI 2025 | Awesome Papers, and I have extracted parts that are of interest.
Model Serving
- SuperServe: Fine-Grained Inference Serving for Unpredictable Workloads [Paper]
- GaTech & UC Berkeley & Adobe
Resource Management
- Granular Management
- Quicksand: Harnessing Stranded Datacenter Resources with Granular Computing [Paper]
- MIT & Brown & USC & VMware Research
- Provide developers with familiar, high-level abstractions (e.g., data structures, batch computing); decompose them into resource proclets, granular units that each primarily consume resources of one type; split, merge, and migrate resource proclets in milliseconds.
- MIT & Brown & USC & VMware Research
- GRANNY: Granular Management of Compute-Intensive Applications in the Cloud [Paper]
- ICL
- Quicksand: Harnessing Stranded Datacenter Resources with Granular Computing [Paper]
- Resource Scheduling
- GREEN: Carbon-efficient Resource Scheduling for Machine Learning Clusters [Paper]
- HKUST
- GREEN: Carbon-efficient Resource Scheduling for Machine Learning Clusters [Paper]
- Serverless Computing
- Making Serverless Pay-For-Use a Reality with Leopard [Paper]
- UW-Madison
- Making Serverless Pay-For-Use a Reality with Leopard [Paper]
- Userspace Scheduling
- The Benefits and Limitations of User Interrupts for Preemptive Userspace Scheduling [Paper]
- UCSD
- 利用Intel新特性用户中断
- The Benefits and Limitations of User Interrupts for Preemptive Userspace Scheduling [Paper]