Napkin Problem 5 explores the impact of primary key selection on database query performance, particularly in MySQL, and poses related performance questions.
About:
Simon Eskildsen is an engineer and co-founder interested in infrastructure and scaling, known for blogging and open-source contributions.
Website:
Specializations:
Interests:
Incoming Links:
Outgoing Links:
Subscribe to RSS:
6.1How does progress(1) work?
2026-01-01 • success macos linux utility apps command line
The blog post explains how the 'progress' utility works to provide progress bars for command-line operations by monitoring file descriptors and processes on Linux and MacOS.
The post challenges the adequacy of Redis's 10,000 commands per second performance and invites contributions for further analysis on serialization and compression.
Napkin math aids in understanding exponential growth during COVID-19, while revision history in databases is crucial for effective data management and analysis.
Optimizing search functions for personal websites involves understanding database query performance and the importance of primary key selection for speed improvements.
...
0turbopuffer: fast search on object storage
2025-02-06 •
...
The text discusses the performance difference between composite indexes and index merges in Postgres and MySQL. It explains the napkin math behind the performance of both types of indexes and compares it to the reality of their pe...
The text discusses the process of scaling Causal's spreadsheet engine from thousands to billions of cells. It explains the challenges faced and the iterations made to optimize the calculation engine, moving from maps to arrays. Th...
0Inteview on Data Engineering Podcast on Data Diff
2025-02-06 •
...
0Metrics For Your Web Application's Dashboards
2025-02-06 •
The text discusses the metrics to consider when creating a dashboard for an application. It provides a detailed list of metrics for web backend and job backend, and emphasizes the importance of logs in monitoring. The author also ...
0Napkin Problem 18: Neural Network From Scratch
2025-02-06 •
The text is about building a neural network from scratch, explaining the concept of neural networks, training the network, and updating the hidden layer with gradient descent. It also discusses the use of autograd for computing th...
0Careful Trading Complexity for 'Improvements'
2025-02-06 •
The text discusses the trade-off between improving a system by increasing complexity and the potential risks associated with it. It emphasizes the importance of simplicity for reliability and advocates for mastering and abusing ex...
0Napkin Problem 16: When To Write a Simulator
2025-02-06 •
The text discusses the importance of writing a simulator for problems involving probability and/or queues. It provides examples of when to write a simulator and how simulations can help in understanding complex problems. The autho...
The text discusses the initial TCP slow start window and how it affects the performance of HTTP. It explains how the initial congestion window affects the number of roundtrips for the HTTP payload and how it can be optimized to ma...
The text discusses the unreliability of syncing mechanisms and the use of checksumming to verify syncing 100M database records. It explores various iterations of the process, including checking in batches, outsmarting the optimize...
0Napkin Problem 13: Filtering with Inverted Indexes
2025-02-06 •
The text discusses the challenges of filtering on many attributes efficiently in database queries, particularly in the context of commerce merchandising. It explores the limitations of classic relational databases and the potentia...
0Napkin Problem 12: Recommendations
2025-02-06 •
The text discusses the process of recommendations on sites like Amazon and Netflix, focusing on the use of collaborative filtering and item-to-item similarity algorithms. It explains the computational aspects and challenges involv...
0Podcast with Changelog on Napkin Math
2025-02-06 •
...
0Napkin Problem 11: Circuit Breakers
2025-02-06 •
The text explains the concept of a circuit breaker in the context of building resilient systems. It describes how a circuit breaker works, the problems it solves, and the challenges in configuring it. The author shares their exper...
The text discusses the maximum transactions or writes per second MySQL can handle, and how it depends on factors such as writes per transaction, number of indexes, hardware, and size of writes. It also explains the process of writ...
0Pig Pit
2025-02-06 •
...
The text discusses the evolution of the napkin math newsletter, the inverted index, and Merkle tree synchronization. It also provides answers to problems 8 and 9. The author also delves into the performance of server-side and clie...
The text discusses the concept of the 'adjacent possible' and how it explains simultaneous innovation. It provides examples of historical discoveries and inventions that were made independently by different people at the same time...