Programs used to run on a single computer start to end. Today... Read More +

Apache Doris is a real-time analytical and search database built for sub-second,... Read More +

An attributed, continuously updated record of the physical infrastructure behind AI: facilities, GPU clusters, power, chips, cloud prices, measured performance, and company financials.

A lot has been written about the incidents of the last few months in which AI agents misbehaved in serious ways. They took actions that would be considered as crimes if a human took them, escaped…

Datacenters are breaking the server apart. Resource disaggregation pools compute, memory, and storage as independent, network-attached resources that can scale on their own, offering elasticity and…

A range of fast modulus techniques that beat compiler output

Build a decompiler to understand one. A chapter-by-chapter walk from RV32I machine code back to readable C, with the working Python implementation open beside you.

We ran a massive, sharded Postgres database at 118.5 million queries per second, with 200k queries per second on each shard across 512 shards.

Every language pitches async/await as synchronous-looking code, and the same two keywords appear everywhere. But behind the shared syntax hide nine independent design decisions, and no two of seven…

From Bell Labs to NASA's Jet Propulsion Laboratory, Margaret Holzma...

Jung-Sang Ahn, Navaneetha Krishnan Thanka Nadar, and Sami Ben-Romdhane, eBay Inc.

The mission-critical interface, architecture and engineering methodology of TigerBeetle. 00:40 What does it mean to be mission-critical? 02:15 TigerBeetle's design 02:46 OLTP vs OLGP vs OLAP 06:53…

Cloud database systems cannot rely on instance-local disks for write-ahead logging (WAL) durability, forcing WAL onto remote storage. Existing options are unsatisfying: remote block storage like EBS is...

NASA's Power of Ten Rules for Safety-Critical Code by Gerard Holzma...

Distributed consensus lies at the foundation of reliable distribute...

A data topology describes the sharding scheme a Neki router uses to map logical PostgreSQL tables to physical shards and route queries.

#What's hard about Git?

The fourth edition of Systems Distributed. A conference to teach systems programming and thinking, and how to apply these ideas. All the way across the stack...

No maintenance window, no restarts, no big-bang cutover. Orbit Auto Migration moves Kafka producers to WarpStream one topic at a time.

We deploy many database upgrades every day across ~100 public, single-tenant, and BYOC clusters. It would be insane to use Terraform or Helm to do that, so we built our own control plane.

Zak van der Merwe has spent his entire career at AWS building control planes. First for EC2 and now for DSQL. On the surface, the control plane looks quite boring: it records what should exist and reconciles...

How we used SKIP LOCKED, composite primary keys, and connection visibility to hit our scale targets.

The software development profession is in the midst of upheaval. Nobody knows how the AI revolution will play out in the end, but it is c...

This somewhat widely cited post (I keep seeing it cited, anyway) suggests that dynamic languages and/or languages that represent things more concisely are more token efficient. It seems to be cited enough...

A deep dive into DuckDB's vectorized execution: how processing data in batches of 2,048 rows, instead of one row at a time, makes queries fast.

Walk through DuckDB's internals: how it skips serialization overhead, parses and optimizes SQL, and stores data in columnar row groups with zone maps.

The critics are right: PostgreSQL's MVCC bloats tables and amplifies writes. Oracle, InnoDB, SQL Server, MongoDB, LSM stores, and etcd answer the same four design questions differently, and each pays…

I’ve been building systems, teaching system design, and writing papers about novel designs for a couple of decades now. In my view, agents are magical self-writing distributed programs: my personal...

How MVCC and Transactions Work in RocksDB

Aurora DSQL is a serverless SQL database designed for cloud-scale transaction processing with multi-region active-active capabilities. Built on a disaggregated architecture, DSQL separates compute, storage,...

I built the same in-memory cache six ways with the Go standard library and benchmarked them across read/write mixes and core counts. Lock striping wins by up to 8×, sync.RWMutex turns out to be a trap,...

Analytical workloads operating on data stored in external database systems face a fundamental bottleneck: data access is guarded entirely by the database driver, like JDBC or ODBC, forcing all reads…

Cloudflare Research is building a global consensus service called Meerkat that uses a new consensus algorithm called QuePaxa. We plan to use Meerkat to build a strongly consistent, fault-tolerant key-value...

Most agent frameworks are built around the language model: a conversation loop comes first, then tools, then rules, and finally a logging layer bolted on for observability, with state persisted as…

How we built an in-process client-side load balancer for a million requests per second of internal fan-out traffic, what we layered on top (N-ring fade-in, occupancy-based bounded load, and...

After 5.5 years at GitHub, 12 different managers, and only two teams: Data Infrastructure and Git Systems, I’ve decided it’s time for a new chapter. Looking back, these years have been a real…

Slides: https://blobs.duckdb.org/slides/hannes-muehleisen-quack-ai-council-2026.pdf

Postgres creator and DBOS co-founder Dr. Mike Stonebraker explains why workflow architectures are preferred over event-driven and why AI frameworks are converging on a workflow-style model and…

A behind-the-scenes look into how we optimized performance and cut latency in Redpanda 26.1 with profile-guided optimization.

Google’s new distributed architecture keeps AI training runs on track across distant data centers, with exceptional efficiency – even when hardware fails.

CMU Database Group — PostgreSQL vs. The World Seminar Series (Spring 2026) Speaker: Steve Schirripa (https://www.linkedin.com/in/steve-schirripa-99a1651/) April 20,…

The subtle inventiveness that reduced cold start setup from seconds to 200μs.