Reading List

Papers and references for the Advanced Topics lectures (Weeks 9–14).

Topics

• Heterogeneous Computing in Cloud — Lec 16–17
• Modern Resource Scheduling — Lec 18–19
• ML in Cloud — Lec 20–21
• Cloud and IoT — Lec 22–23
• Cloud Security — Lec 24
• Future Cloud Infrastructure — Lec 25
• Cloud in the Agentic-AI Era — Lec 26

Heterogeneous Computing in Cloud

Lec 16–17 · 10/20 & 10/22

Lec 16

Lecture:

• Caulfield, Adrian M., et al. "A cloud-scale acceleration architecture." 2016 49th Annual IEEE/ACM international symposium on microarchitecture (MICRO). IEEE, 2016.
• NVIDIA Multi-Instance GPU User guide.

Student Presentation:

• Khawaja, Ahmed, et al. "Sharing, protection, and compatibility for reconfigurable fabric with AmorphOS." OSDI 18.
• Jeon, Myeongjae, et al. "Analysis of Large-Scale Multi-Tenant GPU clusters for DNN training workloads." USENIX ATC 19.

Lec 17

Lecture:

• Li, Suyi, et al. "Heterogeneity at Hyperscale: Characterization and Scheduling of Large Production AI Clusters at Alibaba (Operational Systems)." OSDI 26.
• Yang, Lingyun, et al. "GPU-Disaggregated serving for deep learning recommendation models at scale." NSDI 25.

Student Presentation:

• Chen, Jiyang, et al. "μShell: A Microkernel-based FPGA Shell Architecture." OSDI 26.
• He, Congjie, et al. "WaferLLM: Large language model inference at wafer scale." OSDI 25.

Optional Reading:

• Hei, Chenyang, et al. "HeteCCL: Synthesizing Near-Optimal Collective Communication Schedules for Heterogeneous GPU Clusters." NSDI 26.

Modern Resource Scheduling

Lec 18–19 · 10/27 & 10/29

Lec 18

Lecture:

• Tang, Chunqiang, et al. "Twine: A unified cluster management system for shared infrastructure." OSDI 20.
• Rzadca, Krzysztof, et al. "Autopilot: workload autoscaling at google." EuroSys 20.

Student Presentation:

• Cortez, Eli, et al. "Resource central: Understanding and predicting workloads for improved resource management in large cloud platforms." SOSP 17.

Optional Reading:

• Gog, Ionel, et al. "Firmament: Fast, centralized cluster scheduling at scale." OSDI 16.

Lec 19

Lecture:

• Chai, Xiaohu, et al. "Fork in the road: Reflections and optimizations for cold start latency in production serverless systems." OSDI 25.
• Zhang, Zhengtong, et al. "DVLA: Dynamic VM Lifetime Aware Scheduling for Drifting Lifetime Distributions and Long-Lived VM Placement Debt (Operational Systems)." OSDI 26.

Student Presentation:

• Fu, Yuqi, et al. "ALPS: An Adaptive Learning, Priority OS Scheduler for Serverless Functions." ATC 24.

Optional Reading:

• Liu, Qingyuan, et al. "Harmonizing efficiency and practicability: Optimizing resource utilization in serverless computing with jiagu." ATC 24.

ML in Cloud

Lec 20–21 · 11/3 & 11/5

Lec 20

Lecture:

• Qiao, Aurick, et al. "Pollux: Co-adaptive cluster scheduling for goodput-optimized deep learning." OSDI 21.
• Xiao, Wencong, et al. "Gandiva: Introspective cluster scheduling for deep learning." OSDI 18.

Student Presentation:

• Xiao, Wencong, et al. "AntMan: Dynamic scaling on GPU clusters for deep learning." OSDI 20.

Optional Reading:

• Weng, Qizhen, et al. "MLaaS in the wild: Workload analysis and scheduling in Large-Scale heterogeneous GPU clusters." NSDI 22.

Lec 21

Lecture:

• Xie, Zhiqiang, et al. "Strata: Hierarchical context caching for long context language model serving." OSDI 26.
• Kwon, Woosuk, et al. "Efficient memory management for large language model serving with paged attention." SOSP 23.

Student Presentation:

• Wu, Bingyang, et al. "FastServe: Iteration-Level Preemptive Scheduling for Large Language Model Inference." NSDI 26.

Optional Reading:

• Khare, Alind, et al. "SuperServe: Fine-Grained Inference Serving for Unpredictable Workloads." NSDI 25.

Cloud and IoT

Lec 22–23 · 11/10 & 11/12

Lec 22

Lecture:

• Cuervo, Eduardo, et al. "Maui: making smartphones last longer with code offload." MobiSys 10.
• Gordon, Mark S., et al. "COMET: Code offload by migrating execution transparently." OSDI 12.

Student Presentation:

• Chun, Byung-Gon, et al. "Clonecloud: elastic execution between mobile device and cloud." EuroSys 11.

Optional Reading:

• Cuervo, Eduardo, et al. "Kahawai: High-quality mobile gaming using gpu offload." MobiSys 15.

Lec 23

Lecture:

• Kang, Yiping, et al. 2017. "Neurosurgeon: Collaborative Intelligence Between the Cloud and Mobile Edge." ASPLOS 17.
• Shen, Zheyu, et al. "Edgelora: An efficient multi-tenant llm serving system on edge devices." MobiSys 25.

Student Presentation:

• Prakash, Shvetank, et al. "Lifetime-Aware Design for Item-Level Intelligence at the Extreme Edge." ASPLOS 26.

Optional Reading:

• Sen, Tanmoy, Haiying Shen, and Anand Padmanabha Iyer. "Flex: Fast, accurate DNN inference on low-cost edges using heterogeneous accelerator execution." EuroSys 25.

Cloud Security

Lec 24 · 11/17

Lec 24

Lecture:

• Ristenpart, Thomas, et al. "Hey, you, get off of my cloud: exploring information leakage in third-party compute clouds." CCS 09.
• Baumann, Andrew, et al. "Shielding Applications from an Untrusted Cloud with Haven." OSDI 14.
• Shi, Jiacheng, et al. "Serverless functions made confidential and efficient with split containers." USENIX Security 25.

Student Presentation:

• Zhao, Zirui, et al. "Everywhere all at once: Co-location attacks on public cloud FaaS." ASPLOS 24.

Optional Reading:

• Wang, Shiwen, et al. "Enjoy the Free Lunch, Someone Paid for Us: Escaping Resource Limits of MicroVM-based Containers." USENIX Security 26.
• Liu, Xunqi, et al. "The Dark Side of Flexibility: Detecting Risky Permission Chaining Attacks in Serverless Applications." NDSS 26.

Future Cloud Infrastructure

Lec 25 · 11/19

Lec 25

Lecture:

• Kuchler, Tom, et al. "Unlocking true elasticity for the cloud-native era with dandelion." SOSP 25.
• Ruan, Zhenyuan, et al. "Quicksand: Harnessing stranded datacenter resources with granular computing." NSDI 25.
• Li, Quanxi, et al. "Beehive: A scalable disaggregated memory runtime exploiting asynchrony of multithreaded programs." NSDI 25.

Student Presentation:

• Yi, Shushu, et al. "Espresso: Constructing Cost-Efficient CXL JBOF via Inter-SSD Computing Resource Sharing." OSDI 26.

Optional Reading:

• Qiu, Ziyue, et al. "Moirai: Optimizing Placement of Data and Compute in Hybrid Clouds." SOSP 25.
• Lou, Chiheng, et al. "HydraServe: Minimizing Cold Start Latency for Serverless LLM Serving in Public Clouds." NSDI 26.

Cloud in the Agentic-AI Era

Lec 26 · 11/24

Lec 26

Lecture:

• Chaudhry, Gohar Irfan, et al. "Murakkab:Resource-Efficient Agentic Workflow Orchestration in Cloud Platforms." OSDI 26.
• Gao, Wei, et al. "RollArt: Disaggregated Multi-Task Agentic RL Training at Scale." OSDI 26.

Student Presentation:

• Xiang, Dawei, et al. "LLM-as-Scheduler: Agentic Workflow Dynamic Scheduling." ACL 26.
• Liu, Shiyi, et al. "AIMS: Cost-Efficient LLM-Based Agent Deployment in Hybrid Cloud-Edge Environments." EuroSys 26.

Optional Reading:

• Fang, Taosong, et al. "Flashagents: Accelerating multi-agent llm systems via streaming prefill overlap." MLSys 26.