Vol:18 No:8 → HoliPaxos: Towards More Predictable Performance in State Machine Replication vldb.org/pvldb/vol18/p2…

Microservices are just a fancy way of distributing your bugs across multiple servers.

About ten years ago, as I was working week-ends on open source code, often providing free support left and right, an anonymous reader of my blog sent me 250$. They left a message saying that they were using a library that I was maintaining and enjoying my blog. A few months

Microsoft released a No-code AI data analysis tool. It combines drag-and-drop with natural language to transform data and build visualizations without writing a single line of Python code. 100% opensource.

There’s a whole generation of programmers who have not been taught how to make modular software without using a network stack. We used to teach modularity and loose coupling between in-memory modules. SO much faster and without having to adopt network failures.

Something clicked for me last night when thinking of OLAP DBs. Since data is stored column oriented, a table is a lot diff than an OLTP B+Tree based RDBMS table. OLAP I *think* a table is just a collection of columns. New tables kind of like projections, can reuse columns.

Yup and to further this point - an OLAP table not all “rows” have values. It’s inefficient to store nulls in a column oriented DB. You’re trying to maximize IO throughput and AVX processing.

I’m really curious about Bε-Tree (Buffer Tree) based storage engines but there’s so few you can try out.

I am looking for C/C++ engineers with substantial experience in database servers/storage engines

Thinking of starting a Substack where I can ramble a little longer with my thoughts and ideas than I can on Twitter. I kind of miss my blog but it also made me feel like I had to write with great detail and I would stress myself with accuracy even if I was just learning.

Anyone interested? Just curious 😂

Kelly Sommers Oh definitely, with details, would love to read. First post: a recap of the RavenDB saga, with details about what the issues were, culminating with the Jepsen report.

This paper shows how rearchitecting B+trees for SIMD brings multi-fold improvements in query throughput. If you are keen to deep dive into SIMD, this is a highly recommended read - "A High Throughput B+tree for SIMD Architectures". ece.lsu.edu/lpeng/papers/t…

🔥 Postgres 18 update: indexable UUIDs Postgres 18 is in beta and will be out in production databases near you this fall. There’s a new addition to the uuid data type for supporting UUIDv7. UUIDs are randomly generated strings. They are inherently globally unique. UUIDs are

TIL Koen Claessen, the inventor of async await is also the inventor of property-based testing Legend

On this date in 2015, Takuma Sato saw cars going six-wide in front of him and decided to go seven-wide and pass all of them. 12th to 4th in five seconds

actual footage of my PM trying to assign me a jira

What a race! A special one here in Spa ✨🩷 #LMC #IronDames #WomenDrivenByDreams #BeAnIronDame