Introduction
Introduction
This website hosts my notes for papers that I've read.
Disclaimer: These notes shared here are by no means professional nor accurate. I'm simply sharing my thoughts at the exact time when I read this paper with my limited background.
Related Course Notes
CS 245 Principles of Data-Intensive Systems @ Stanford
CSE 599W Systems for ML @ UW
Distributed ML by Tie-Yan Liu (in Chinese)
Databases
Staring into the Abyss: An Evaluation of Concurrency Control with One Thousand Cores
Calvin: Fast Distributed Transactions for Partitioned Database Systems
A Study of the Fundamental Performance Characteristics of GPUs and CPUs for Database Analytics
Q100: The Architecture and Design of a Database Processing Unit
Managing Non-Volatile Memory in Database Systems
Write-Behind Logging
Joins in a Heterogeneous Memory Hierarchy: Exploiting High-Bandwidth Memory
Query Processing on Smart SSDs: Opportunities and Challenges
Database Processing-in-Memory: An Experimental Study
The End of Slow Networks: It's Time for a Redesign
Rethinking Database High Availability with RDMA Networks
Offloading Distributed Applications onto SmartNICs using iPipe
Distributed Join Algorithms on Thousands of Cores
Amazon Aurora: Design Considerations for High Throughput Cloud-Native Relational
Choosing A Cloud DBMS: Architectures and Tradeoffs
The Snowflake Elastic Data Warehouse
Starling: A Scalable Query Engine on Cloud Function Services
HyPer: A Hybrid OLTP&OLAP Main Memory Database System Based on Virtual Memory
Gorilla: A Fast, Scalable, In-Memory Time Series Database
Adaptive Concurrency Control: Despite the Looking Glass, One Concurrency Control Does Not Fit All
Toward Coordination-Free and Reconfigurable Mixed Concurrency Control
To Read List
Distributed Systems
To Read List
ML & Big Data Systems
To Read List
Themis: Fair and Efficient GPU Cluster Scheduling - NSDI 2020
Blink: Fast and Generic Collectives for Distributed ML - MLSys 2020
Accelerating Deep Learning Inference via Freezing - HotCloud 2019
Analysis of Large-Scale Multi-Tenant GPU Clusters for DNN Training Workloads - USENIX ATC 2019
Theory
To Read List
Last updated