| Feb. 18: Introduction |
|
The task of the referee |
|
|---|
| Feb. 25: Datacenter basics |
|
|
Web search for a planet: The Google cluster architecture [Google, IEEE Micro '03] |
|
|---|
|
|
A view of cloud computing [Berkeley, CACM '10] |
|
|---|
| Mar. 04: Datacenter workloads |
|
|
Clearing the Clouds: a study of emerging scale-out workloads on modern hardware [EPFL, ASPLOS '12] |
|
|---|
|
|
DCPerf: An Open-Source, Battle-Tested Performance Benchmark Suite for Datacenter Workloads [Meta, ISCA '25] |
|
|---|
| Mar. 11: Quality of service |
|
|
Tail at Scale [Google, CACM '13] |
|
|---|
|
|
Amdahl’s Law for Tail Latency [Stanford, CACM '18] |
|
|---|
| Mar. 18: Processor design |
|
|
Scale-out Processors [EPFL, ISCA '12] |
|
|---|
|
|
The Sharing Architecture: Sub-Core Configurability for IaaS Clouds [Princeton, ASPLOS '14] |
|
|---|
| Mar. 25: No class |
| Apr. 01: Total cost of ownership |
|
|
ASIC Clouds: Specializing the Datacenter [UWashington, ISCA '16] |
|
|---|
|
|
Moonwalk: NRE Optimization in ASIC Clouds [UWashington, ASPLOS '17] |
|
|---|
| Apr. 08: No class |
| Apr. 15: Power and sustainability |
|
|
Power Provisioning for a Warehouse-sized Computer [Google, ISCA '07] |
|
|---|
|
|
Sustainable AI: Environmental implications, challenges and opportunities [Meta, MLSys '22] |
|
|---|
| Apr. 22: Memory/Storage |
|
|
The case for RAMClouds: scalable high-performance storage entirely in DRAM [Stanford, SIGOPS '10] |
|
|---|
|
|
Disaggregated memory for expansion and sharing in blade servers [Michigan, HP, & AMD, ISCA '09] |
|
|---|
| Apr. 29: Networking |
|
|
Homa: A Receiver-Driven Low-Latency Transport Protocol Using Network Priorities [Stanford & MIT, SIGCOMM '18] |
|
|---|
|
|
RPCValet: NI-Driven Tail-Aware Balancing of µs-Scale RPCs [EPFL, ASPLOS '19] |
|
|---|
| May. 06: Rackscale design |
|
|
Pond: CXL-Based Memory Pooling Systems for Cloud Platforms [Microsoft, ASPLOS '23] |
|
|---|
| May. 13: Serverless computing |
|
|
Architectural Implications of Function-as-a-Service Computing [Princeton, MICRO '19] |
|
|---|
|
|
Single-Address-Space FaaS with Jord [EPFL, Yale, & Technion, ISCA '25] |
|
|---|
| May. 20: Deep learning |
|
|
In-Datacenter Performance Analysis of a Tensor Processing Unit [Google, ISCA '17] |
|
|---|
|
|
The Architectural Implications of Facebook's DNN-based Personalized Recommendation [Meta, HPCA '20] |
|
|---|
| May. 27: LLM systems in datacenters |
|
|
Orca: A Distributed Serving System for Transformer-Based Generative Models [SNU & FriendliAI, SNU '22] |
|
|---|
|
|
Splitwise: Efficient Generative LLM Inference Using Phase Splitting [UWashington & Microsoft, ISCA '24] |
|
|---|
