Google Dapper, a Large-Scale Distributed Systems Tracing Infrastructure
Gaurav Sen
@gkcsAbout
This channel is for software developers who want to learn artificial intelligence, system design, and algorithms. You will find content around LLMs, data structures, algorithms, game-playing AI, and system design.
Latest Posts
Video Description
The dapper system allows real-time request tracing at Google Scale. Here are some of its salient features: 1. It stores 1 TB of request traces per day! 2. Dapper is used by nearly all systems at Google 3. The client libraries are incredibly lightweight, consuming less than 1% of a CPU core. 00:00 What is Google Dapper? 00:53 Request Tracing Requirements 03:59 Data Schema 06:15 High Level Design 09:00 Latency Reduction - Sampling 11:57 Thank you! Dapper Paper: http://research.google/pubs/dapper-a-large-scale-distributed-systems-tracing-infrastructure/ BigTable Paper: https://www.usenix.org/legacy/event/osdi06/tech/chang/chang.pdf Memcached Video: https://youtu.be/uS8f5SSYDck Monarch Video: https://youtu.be/NxPTVhc1tFA New SYSTEM DESIGN series launched here: https://interviewready.io/learn/system-design-course #SystemDesign #Google #ResearchPaper
No Recommendations Found
No products were found for the selected channel.