New blog about MapDB 4 https://t.co/mD9tDiWcpX

I'll be happy to talk to teams that currently work on languages targeting 100K+ users and figure out how I can help. Probably, no full-time commitments, but I'm pretty sure I can be useful in many other ways. 2/2

"How Query Engines Work" Ever wanted to build your own query engine? Then check out this guide by @andygrove_io, touching on all the relevant parts like query parsing, planning, optimizing, etc. https://t.co/tbXWuhuc6J

Looked into HP Dev One laptop more. It has some sort of privacy screen filter, so 1000 nits is not really brighter than normal 400 nits laptop. Also bad battery life. Really disappointing. #HPDevOne

I am looking for Linux laptop, that can do 1000 nits and 64GB RAM. Is there a reasonable way to get HP Dev One in EU? #HPDevOne

Nice talk about Buffer Trees (aka Cola trees, Fractal, B-Epsilon). Congrats to authors for good implementation, but for me this stuff is just to complex. I hope I will be able to replicate some benefits with storage log buffers, on regular BTree.

There is new MapDB release; 3.0.9, new DBMaker.fileSyncDisable() option.

Some design changes to solve write amplification in MapDB 4 https://t.co/oE1OCi7fly

This Monday I will have a talk about MapDB in Vienna: https://t.co/Pg3p4AB7m6

I am looking for work. Senior backend java developer. Experience with Kotlin, Scala, DBs, crypto protocols.. My main OS project is MapDB. Preferably remote in a small team.

I am working on Eclipse Collections version that would use offheap or mmaped ByteBuffers. EC already have many concepts from MapDB4, and is good starting point for new MapDB version.

Aeron 1.29.0 is out! C Media Driver and Cluster get some love plus some performance improvements and bug fixes. https://t.co/lCESTOhpuq

I could use some beefy machine (1TB RAM+) to test mapdb. Just overnight benchmarks via SSH

MapDB update

February update for MapDB https://t.co/9Ya5nJhW1Y

MapDB site is moving from Disqus to github based comment system.

Contrary to common belief, a #Java object's identity hash code is not necessarily related to its memory address. Instead, values are obtained via a configurable strategy. On OpenJDK, by default, it's a random value. Details in this post by @alidehghanig:

https://t.co/PUWfImuxtG now uses https by default. Time to cleanup this project :)

After long break I am back working on MapDB. More details are coming. Happy New Year.

zig related io_uring benchmarks: https://t.co/A2SDsAgToj What is often overlooked, but I believe the most important change, is quoted in there too: "What is unique about io_uring here is that the same simple interface can be used for both file system and networking I/O"