Miguel Ángel Pastor Olivar

The Quest for One Million IOPS: Benchmarking Storage at LanceDB Learn how LanceDB benchmarks storage and how we achieved one million disk reads per second.

The Quest for One Million IOPS: Benchmarking Storage at LanceDB

Clickhouse Internals: A Deep Dive into ClickHouse Distributed Connection Pooling A deep dive into ClickHouse distributed connection pooling and why bursty workloads can trigger scary warnings without real failures.

Clickhouse Internals: A Deep Dive into ClickHouse Distributed Connection Pooling medium.com/@pranavmehta...

CockroachDB Serverless: Sub-second Scaling from Zero withMulti-region Cluster Virtualization dl.acm.org/doi/epdf/10....

The Art of Being Lazy(log): Lower latency and Higher Availability With Delayed Sequencing We're excited to launch Lightning Topics and Ripcord Mode. Lightning Topics use delayed sequencing to achieve 33ms median produce latency with S3 Express One Zone — a 70% reduction. Ripcord Mode lets…

The Art of Being Lazy(log): Lower latency and Higher Availability With Delayed Sequencing www.warpstream.com/blog/the-art...

Complexity in Computer Science Complexity in Computer Science is a textbook on computational complexity theory by Thomas Watson.

Complexity in Computer Science

TigerStyle TigerStyle is a coding philosophy focused on safety, performance, and developer experience.

TigerStyle: The software engineering methodology developed by TigerBeetle to produce safer, faster software in less time. tigerstyle.dev

2026-01-31: Column Storage for the AI Era Joint Seminar (with TU Darmstadt) Column Storage for the AI Era January 31 2026: Joint Database Systems Seminar with TU Darmstadt Joint work with Julien Le Dem (Co-creator of Apache Parquet) Title: "Column Storage for the AI Era",…

Column Storage for the AI Era docs.google.com/presentation...

fosdem.org/2026/events/...

FOSDEM 2026 - From Disks to Distributed: Our Journey of Database Evolution in the Cloud Our database had reached a point where failure scenarios were becoming increasingly complex and time consuming. A single node could take up to 15 minutes to recover. It was expensive to run and…

From Disks to Distributed: Our Journey of Database Evolution in the Cloud fosdem.org/2026/schedul...

What We Talk About When We Talk About Agent Infra me.0xffff.me

XRP: In-Kernel Storage Functions with eBPF

PostgreSQL vs. The World Seminar Series - Carnegie Mellon Database Group Every major cloud vendor now offers an enhanced, opinionated PostgreSQL-compatible database management... Read More +

PostgreSQL vs. The World Seminar Series

GitHub - microsoft/bf-tree: Bf-Tree is a modern read-write-optimized concurrent larger-than-memory range index in Rust from MS Research. Bf-Tree is a modern read-write-optimized concurrent larger-than-memory range index in Rust from MS Research. - microsoft/bf-tree

Bf-Tree: a modern read-write-optimized concurrent larger-than-memory range index in Rust from MS Research.

Formal Methods Beyond Correctness: Isolation & Permissiveness of Distributed Transactions in MongoDB Learn how we used modular protocol specification for verifying both correctness and performance of MongoDB’s distributed transactions protocol.

Formal Methods Beyond Correctness: Isolation & Permissiveness of Distributed Transactions in MongoDB www.mongodb.com/company/blog...

Database Recovery Demystified: Understanding ARIES from First Principles | Yash Agarwal

The Memory Paradox: Why Our Brains Need Knowledge in an Age of AI papers.ssrn.com/sol3/papers....

Scaling PostgreSQL to power 800 million ChatGPT users openai.com/index/scalin...

Semantic Data Modeling, Graph Query, and SQL, Together at Last? storage.googleapis.com/gweb-researc...

Unconventional PostgreSQL Optimizations Creative ideas for speeding up queries in PostgreSQL

Unconventional PostgreSQL Optimizations. Creative ideas for speeding up queries in PostgreSQL hakibenita.com/postgresql-u...

Writing a TSDB from scratch in Go Writing a TSDB from scratch in Go Performance optimizations

Writing a TSDB from scratch in Go docs.google.com/presentation...

GongShow_CIDR.pdf

The End of General-purpose Database Systems? Or why we need to clone! drive.google.com/file/d/1XPQg...

Are Database System Researchers Making Correct Assumptions about Transaction Workloads? In this blog, we had reviewed quite a number of deterministic database papers, including Calvin , SLOG , Detock , which aimed to achieve hig...

Are Database System Researchers Making Correct Assumptions about Transaction Workloads?

Building a Corruption-Proof Write-Ahead Log in Go Learn how to build a crash-safe Write-Ahead Log (WAL) in Go, and why CRC32 alone is not enough. We explore the durability layers UnisonDB uses to prevent corruption after crashes.

Building a Corruption-Proof Write-Ahead Log in Go

Demystifying and Improving Lazy Promotion in Cache Eviction

Pruning in Snowflake: Working Smarter, Not Harder arxiv.org/pdf/2504.11540

Clock Synchronization Is a Nightmare Time seems simple. But we engineers lose sleep over something as basic as keeping clocks in sync. Here’s why…

Clock Synchronization Is a Nightmare

Clock Synchronization Is a Nightmare Time seems simple. But we engineers lose sleep over something as basic as keeping clocks in sync. Here’s why…

Clock Synchronization Is a Nightmare

GitHub - djiangtw/data-structures-in-practice-public: A hardware-aware guide to data structures for system software engineers. A hardware-aware guide to data structures for system software engineers. - djiangtw/data-structures-in-practice-public

A hardware-aware guide to data structures for system software engineers.

Databases in 2025: A Year in Review The world tried to kill Andy off but he had to stay alive to to talk about what happened with databases in 2025.

Databases in 2025: A Year in Review

ArceKV: Towards Workload-driven LSM-compactions for
Key-Value Store Under Dynamic Workloads