# vacuum (blogs) — RSS Amplifier Recent posts from the 4 feeds in the RSS Amplifier directory that cover vacuum. Page: Feed: --- ## [A South Asian eyeful](https://www.eventslondon.org/reviews/a-south-asian-eyeful) _2026-08-09 · Daniel Nelson · Reviews - London listings: events about Africa, Asia & Latin America_ For a South Asian treat and an eyeful of images go now to Christie’s, which is showing a wonderful, eclectic selection of paintings, sculptures and videos from the region. ## [The tug of home, the lure of the street](https://www.eventslondon.org/reviews/the-tug-of-home-the-lure-of-the-street) _2026-07-25 · Daniel Nelson · Reviews - London listings: events about Africa, Asia & Latin America_ It’s street versus home in Ish, a coming of age story about a Muslim British Asian in the Bedfordshire town of Luton. ## [How OOMWOO cleaning algorithms works](https://makerspet.com/blog/oomwoo-phantom-obstacle-cleaning-fix/) _2026-07-22 · Ilia O. · Makers Pet_ In simulation, our open-source robot vacuum kept grinding to a halt in front of the sofa. The obstacle turned out to be a phantom on the costmap. Here is how we found it by measuring, and the two-part fix that roughly doubled coverage near obstacles. ## [OOMWOO Headless Sim for LLM Agents](https://makerspet.com/blog/oomwoo-headless-sim-coverage-cleaning-llm-agents/) _2026-07-22 · Ilia O. · Makers Pet_ A concise, copy-paste quickstart for running the OOMWOO robot vacuum entirely headless (no display) — aimed at LLM coding agents, CI, and automation. Every command below is verified against the makerspet/oomwoo:jazzy-dev Docker image (last re-verified 9 Aug 2026). No X server, GPU, or GUI is required; Gazebo renders offscreen with software GL. 1. Start the \[ \] ## [Phone calls from the Gaza apocalypse ](https://www.eventslondon.org/reviews/phone-calls-from-the-gaza-apocalypsenbsp) _2026-07-21 · Daniel Nelson · Reviews - London listings: events about Africa, Asia & Latin America_ Life Support is a doctors’-eye-view of the Israeli attack on Gaza. And what a visceral, appalling view it is. ## [Israel is off the rails, says brain-fry documentary](https://www.eventslondon.org/reviews/4ei9ea1leszrg3uf6umji2sppyn43h) _2026-07-21 · Daniel Nelson · Reviews - London listings: events about Africa, Asia & Latin America_ ‘This film will fry your brain’, proclaims a new documentary, Planet Israel. ## [Photographer of protest](https://www.eventslondon.org/reviews/photographer-of-protest) _2026-07-04 · Daniel Nelson · Reviews - London listings: events about Africa, Asia & Latin America_ I was in two minds about the new biopic of Nigerian-British photographer Misan Harriman — exciting shots of political demonstrations but is it a little too adulatory? — until I read about right-wing cultural attacks on him. ## [Write Your First OOMWOO ROS 2 Package: Cover the Floor While Mapping](https://makerspet.com/blog/write-your-first-oomwoo-ros-2-package/) _2026-07-01 · Ilia O. · Makers Pet_ This is the hello world of developing for OOMWOO. It s deliberately simple — a reactive drive forward, turn when blocked coverage that, combined with SLAM, maps and roughly covers a room. Proper boustrophedon coverage is the clean-and-map RFC; this teaches you the mechanics of writing and launching an OOMWOO node. More tutorials in this series \[ \] ## [Simulate OOMWOO-One Robot Vacuum in Gazebo with ROS 2](https://makerspet.com/blog/simulate-oomwoo-one-robot-vacuum-in-gazebo-with-ros-2/) _2026-07-01 · Ilia O. · Makers Pet_ OOMWOO is an open-source robot vacuum you build yourself. oomwoo-one is the first model. This tutorial gets its ROS 2 simulation running in Gazebo so you can develop mapping, navigation, and cleaning behaviours with no hardware — everything runs in Docker on Ubuntu or Windows. You ll get: SLAM mapping, autonomous Nav2 navigation, manual driving, and \[ \] ## [How-to: Source BOM for OOMWOO Open-Source Vacuum Robot](https://makerspet.com/blog/how-to-source-bom-for-oomwoo-open-source-vacuum-robot/) _2026-06-30 · Ilia O. · Makers Pet_ I started by drafting a list of parts and deciding which have to be sourced and which should be 3D printed. I searched each part on AliExpress to see what is available abundantly and at what price. Find the master Bill of Materials (BoM) here. Caution this post may contain errors and \[ \] ## [CLI-first decentralized GPU compute (Sponsored)](https://crawlproof.com/a/mvN27wySQSYA) _2026-06-30 · **Sponsored**_ Pay workers or run your GPU for FFmpeg transcode and AI inference. ## [How the Zeitgeist helped create Fridamania](https://www.eventslondon.org/reviews/how-the-zeitgeist-helped-create-fridamania) _2026-06-23 · Daniel Nelson · Reviews - London listings: events about Africa, Asia & Latin America_ Advance sales for the new Frida Kahlo exhibition topped 41,000, a Tate Modern record. At £25 a pop, it will be a commercial bonanza. ## [Tutorial Part 2: Drive, Map & Navigate Your Proscenic M6 Pro in ROS 2](https://makerspet.com/blog/tutorial-part-2-drive-map-navigate-your-proscenic-m6-pro-in-ros-2/) _2026-06-22 · Ilia O. · Makers Pet_ Tutorial Part 1 got SangamIO running on the vacuum, exposing its motors, wheel encoders, IMU and LiDAR over TCP port 5555. Part 2 connects that to ROS 2 the easy way — a prebuilt Docker image that already bundles everything: ROS 2 Jazzy, the vacuum\_ros2\_bridge, the proscenic\_m6pro robot description, an EKF for odometry, cartographer SLAM, \[ \] ## [From Israel to Canada: yesterday’s soldier looks back with regret](https://www.eventslondon.org/reviews/from-israel-to-canada-yesterdays-soldier-looks-back-with-regret) _2026-06-20 · Daniel Nelson · Reviews - London listings: events about Africa, Asia & Latin America_ What a tragically sad title for Itai Erdal’s monologue at the Finborough Theatre: Soldiers of Tomorrow. ## [CPU-bound sysbench on a large server: Postgres 12 to 19 beta1](http://smalldatum.blogspot.com/2026/06/cpu-bound-sysbench-on-large-server.html) _2026-06-20 · Mark Callaghan · Small Datum_ This has results from sysbench on a small server with Postgres versions 12 through 19 beta1. Sysbench is run with high concurrency (40 connections) and a cached database. The purpose is to search for changes in performance. Postgres remains boring, it is hard to find performance regressions. tl;dr for Postgres 17 to 19 there are no regressions throughput on the read-only-count test improves by ~3X… ## [When the aid worker met the Kenyan activist](https://www.eventslondon.org/reviews/when-the-aid-worker-met-the-kenyan-activist) _2026-06-19 · Daniel Nelson · Reviews - London listings: events about Africa, Asia & Latin America_ What has Florence Nightingale got to do with the post-World War Two international development aid industry? ## [A Fine Idea joins the dots on the failure of aid](https://www.eventslondon.org/reviews/bvfht39pkx0vdjufp40fhljx412zmc) _2026-06-19 · Daniel Nelson · Reviews - London listings: events about Africa, Asia & Latin America_ Theatre should make you think, but the Arcola is about to ask audiences to consider a particularly awkward and painful question: if you support international aid, do you really want to change the world — or do you just like the idea of helping? ## [Simulate the Proscenic M6 Pro robot vacuum in Gazebo with ROS 2](https://makerspet.com/blog/simulate-the-proscenic-m6-pro-robot-vacuum-in-gazebo-with-ros-2/) _2026-06-19 · Ilia O. · Makers Pet_ Run a full Proscenic M6 Pro vacuum simulation in Gazebo with the oomwoo ROS 2 image: SLAM mapping, autonomous Nav2 navigation, manual driving and bumper sensors no hardware needed. ## [(Machine) Consciousness](https://linas.org/blog/2026/06/machine-consciousness/) _2026-06-17 · linas · Blog Linas_ Recently, I have been thinking about what it feels like (roughly, Heidegger s Dasein) and sensory perception. I feel like I made a big break-through when I realized that standard emotions (love, lust, loneliness, jealousy ) are kind-of uninvited guests that show up whenever they feel like, and wreak havoc with your life. So my thoughts turned to \[ \] ## [The insert benchmark on a small server, IO-bound workload : Postgres 19 beta1](http://smalldatum.blogspot.com/2026/06/the-insert-benchmark-on-small-server-io.html) _2026-06-17 · Mark Callaghan · Small Datum_ This has results for Postgres versions 19 beta1, 18.4 and 17.10 with the Insert Benchmark on a small server using a cached and CPU-bound workload. I also used MySQL 8.4.8 to see where performance was different. Postgres continues to be boring in a good way. It is hard to find performance regressions. tl;dr create index (the l.x step) is faster in Postgres 19beta1. A Postgres expert told me that… ## [Tutorial: Connect Robot Vacuum Cleaner to ROS 2 – Proscenic M6 Pro](https://makerspet.com/blog/tutorial-connect-robot-vacuum-cleaner-to-ros-2-proscenic-m6-pro/) _2026-06-15 · Ilia O. · Makers Pet_ Turn a Proscenic M6 Pro robot vacuum into a ROS 2 robot: root it over ADB, flash the open-source SangamIO firmware, and bridge its LiDAR, odometry and motors into ROS 2 Jazzy as /scan, /odom and /cmd\_vel. ## [Ship software peacefully (Sponsored)](https://crawlproof.com/a/yosQRYHX4CM6) _2026-06-15 · **Sponsored**_ Connect your repo and deploy instantly — Railway handles config, scaling, and monitoring. ## [Building an Open-Source Robot Vacuum — Meet OOMWOO](https://makerspet.com/blog/building-an-open-source-robot-vacuum-meet-oomwoo/) _2026-06-15 · Ilia O. · Makers Pet_ I’m starting a new build-in-public project: oomwoo, an open-source robot vacuum you build yourself. Raspberry Pi, ROS 2, 2D LiDAR, Home Assistant, 3D printed, local-first — and open from the first commit. ## [HammerDB tproc-c on a large server, Postgres 14 to 19 beta1](http://smalldatum.blogspot.com/2026/06/hammerdb-tproc-c-on-large-server.html) _2026-06-13 · Mark Callaghan · Small Datum_ This has results for HammerDB tproc-c on a large server using MySQL and Postgres. I am new to HammerDB and still figuring out how to explain and present results so I will keep this simple and just share graphs without explaining the results. tl;dr There are small regressions in versions 16, 17 and 18 NOPM usually improves a small amount in 19 beta1 relative to 18 Builds, configuration and hardware… ## [A hen, a ghost in a vacuum cleaner and now talking canids](https://www.eventslondon.org/reviews/a-hen-a-ghost-in-a-vacuum-cleaner-and-now-talking-canids) _2026-06-12 · Daniel Nelson · Reviews - London listings: events about Africa, Asia & Latin America_ Films are getting odder. After a Thai drama about ghosts who inhabit vacuum cleaners (A Useful Ghost) and a people-smuggling drama seen through the eyes of a chicken (Hen), we now have Bouchra, a Moroccan mother-daughter movie acted by canids. ## [Write-heavy sysbench tests, a large server, modern Postgres and MySQL](http://smalldatum.blogspot.com/2026/06/write-heavy-sysbench-tests-large-server.html) _2026-06-12 · Mark Callaghan · Small Datum_ This has results for modern Postgres and MySQL using write-heavy tests from sysbench and a large server. I think there are regressions in Postgres that arrive in some of versions 16, 17, 18 and 19 beta1 but I am far from certain and this blog post is just another step in my journey to figure that out. tl;dr Postgres suffers a lot from throughput variation while MySQL+InnoDB does not InnoDB gets… ## [The insert benchmark on a small server, cached workload : Postgres 19 beta1](http://smalldatum.blogspot.com/2026/06/the-insert-benchmark-on-small-server.html) _2026-06-11 · Mark Callaghan · Small Datum_ This has results for Postgres versions 19 beta1, 18.4 and 17.10 with the Insert Benchmark on a small server using a cached and CPU-bound workload. Postgres continues to be boring in a good way. It is hard to find performance regressions. tl;dr I don't see regressions here in 19 beta1 I see some improvements here in 19 beta1 index create (l.x) is faster but the step is short-running so I don't… ## [Postgres 19 beta1 vs sysbench on a small server](http://smalldatum.blogspot.com/2026/06/postgres-19-beta1-vs-sysbench-on-small.html) _2026-06-10 · Mark Callaghan · Small Datum_ This has results from sysbench on a small server with Postgres 19 beta1, 18.4 and 17.10. Sysbench is run with low concurrency (1 thread) and a cached database. The purpose is to search for changes in performance, often from new CPU overheads. tl;dr 19beta1, 18.4 and 17.10 have mostly similar performance There might be small regressions (about 2%) from 17.10 to 19beta1 but my tests are not good at… ## [lds2d: One Python library for 2D LiDARs](https://makerspet.com/blog/lds2d-python-2d-lidar-library-live-browser-radar/) _2026-06-03 · Ilia O. · Makers Pet_ lds2d is a pure-Python driver for 2D LiDARs on Linux including the Raspberry Pi — a Pythonic port of the kaiaai/LDS C++ library. lds2d supports 23+ LiDAR models today: Try it in 60 seconds without any hardware Once lds2d launches in demo mode, open http://localhost:8080 in your browser to view live synthetic data. Usage \[ \] ## [AI and Regulations](https://linas.org/blog/2026/05/ai-and-regulations/) _2026-05-02 · linas · Blog Linas_ During a recent debate regarding Chinese and US AI geopolitical strategy, and how AI could be, should be integrated into society, a discussion of disconnects arose. The one thing popped into focus as I read through was this: the disconnect between regulatory policy and AI. Let me give an example by copying text from a \[ \] ## [The Physics of Here-and-Now](https://linas.org/blog/2026/04/the-physics-of-here-and-now/) _2026-04-25 · linas · Blog Linas_ Pursuant to the emails Delerium as a Phase Transition and Quantum Vacuum Fluctuations, there was another remark. My reply: Hi Anton! On Tue, Apr 21, 2026 at 9:39 PM Anton Kolonin @ Gmail a @gmail.com\> wrote: Welcome to the conference The Evolution of Complexity and Statistical Physics. Wow! That’s quite the table of contents! Statistical physics overlaps \[ \] ## [Quantum Vacuum Fluctuations](https://linas.org/blog/2026/04/quantum-vacuum-fluctuations/) _2026-04-25 · linas · Blog Linas_ During an email exchange, captured in the previous blog post Delerium as a Phase Transition , someone wrote back. I responded compulsively: On Tue, Apr 21, 2026 at 6:50 PM Keith Henson h @gmail.com\> wrote: Good essay on the nonsense of our times. But you never got around to saying vacuum energy is nonsense. Oh, Sorry. It’s nonsense–ish–asterisk. \[ \] ## [ThreatCrush on GitHub (Sponsored)](https://crawlproof.com/a/42cADtsue7M6) _2026-04-25 · **Sponsored**_ View the ThreatCrush listing on the GitHub Marketplace. ## [Delerium as a Phase Transition of the Global Brain](https://linas.org/blog/2026/04/delerium-as-a-phase-transition-of-the-global-brain/) _2026-04-25 · linas · Blog Linas_ Where, exactly, does the present-day spate of crackpot ideas come from? I claim that social media has fundamentally rewired how how the Global Brain thinks. I claim it is due to a phase transition, in the sense of statistical mechanics. I replied in an email: Well, given that this is the Lifeboat mailing list, I \[ \] ## [MySQL 9.7.0 vs sysbench on a small server](http://smalldatum.blogspot.com/2026/04/mysql-970-vs-sysbench-on-small-server.html) _2026-04-10 · Mark Callaghan · Small Datum_ This has results from sysbench on a small server with MySQL 9.7.0 and 8.4.8. Sysbench is run with low concurrency (1 thread) and a cached database. The purpose is to search for changes in performance, often from new CPU overheads. I tested MySQL 9.7.0 with and without the hypergraph optimizer enabled. I don't expect it to help much because the queries run here are simple. I hope to learn it… ## [Sysbench vs MySQL on a small server: another way to view the regressions](http://smalldatum.blogspot.com/2026/04/sysbench-vs-mysql-on-small-server-n.html) _2026-04-09 · Mark Callaghan · Small Datum_ This post provides another way to see the performance regressions in MySQL from versions 5.6 to 9.7. It complements what I shared in a recent post . The workload here is cached by InnoDB and my focus is on regressions from new CPU overheads. The good news is that there are few regressions after 8.0. The bad news is that there were many prior to that and these are unlikely to be undone. tl;dr for… ## [The Insert Benchmark vs MariaDB 10.2 to 13.0 on a 32-core server](http://smalldatum.blogspot.com/2026/04/the-insert-benchmark-vs-mariadb-102-to_8.html) _2026-04-08 · Mark Callaghan · Small Datum_ This has results for MariaDB versions 10.2 through 13.0 vs the Insert Benchmark on a 32-core server. The goal is to see how performance changes over time to find regressions or highlight improvements. My previous post has results from a 24-core server. Differences between these servers include: RAM - 32-core server has 128G, 24-core server has 64G fsync latency - 32-core has an SSD with high fsync… ## [The Insert Benchmark vs MariaDB 10.2 to 13.0 on a 24-core server](http://smalldatum.blogspot.com/2026/04/the-insert-benchmark-vs-mariadb-102-to.html) _2026-04-08 · Mark Callaghan · Small Datum_ This has results for MariaDB versions 10.2 through 13.0 vs the Insert Benchmark on a 24-core server. The goal is to see how performance changes over time to find regressions or highlight improvements. MariaDB 13.0.0 is faster than 10.2.30 on most benchmark steps and otherwise as fast as 10.2.30. This is a great result. tl;dr for a CPU-bound workload the write-heavy steps are much faster in 13.0.0… ## [Sysbench vs MariaDB on a small server: using the same charset for all versions](http://smalldatum.blogspot.com/2026/04/sysbench-vs-mariadb-on-small-server.html) _2026-04-06 · Mark Callaghan · Small Datum_ This has results for sysbench vs MariaDB on a small server. I repeated tests using the same charset (latin1) for all versions as explained here . In previous results I used a multi-byte charset for modern MariaDB (probably 11.4+) by mistake and that adds a 5% CPU overhead for many tests. tl;dr MariaDB has done much better than MySQL at avoid regressions from code bloat. There are several… ## [CPU-bound sysbench on a large server: Postgres, MySQL and MariaDB](http://smalldatum.blogspot.com/2026/04/cpu-bound-sysbench-on-large-server.html) _2026-04-04 · Mark Callaghan · Small Datum_ This post has results for CPU-bound sysbench vs Postgres, MySQL and MariaDB on a large server using older and newer releases. The goal is to measure: how performance changes over time from old versions to new versions performance between modern MySQL, MariaDB and Postgres The context here is a collection of microbenchmarks using a large server with high concurrency. Results on other workloads… ## [gcc vs clang for sysbench on a small server with Postgres, MySQL and MariaDB](http://smalldatum.blogspot.com/2026/03/gcc-vs-clang-for-sysbench-on-small.html) _2026-03-31 · Mark Callaghan · Small Datum_ This has results for sysbench on a small server and compares performanc for Postgres, MySQL and MariaDB compiled using clang vs using gcc. tl;dr Throughput with clang and gcc is similar Builds, configuration and hardware I compiled Postgres 18.3, MySQL 8.4.8 and MariaDB 11.8.6 from source. The server has 8 AMD cores with SMT disabled and 32G of RAM. The OS is Ubuntu 24.04, gcc is version 13.3.0… ## [The insert benchmark on a small server : Postgres 12.22 through 18.3](http://smalldatum.blogspot.com/2026/03/the-insert-benchmark-on-small-server.html) _2026-03-29 · Mark Callaghan · Small Datum_ This has results for Postgres versions 12.22 through 18.3 with the Insert Benchmark on a small server. My previous post for the same hardware with results up to Postgres 18.1 is here . This post also has results for: all 17.x releases from 17.0 through 17.9 18.2 with and without full page writes enabled both 1 and 4 users Postgres continues to be boring in a good way. It is hard to find… ## [Selecting a character set for MySQL and MariaDB clients](http://smalldatum.blogspot.com/2026/03/selecting-character-set-for-mysql-and.html) _2026-03-29 · Mark Callaghan · Small Datum_ MySQL and MariaDB have many character-set related options, perhaps too many: character\_set\_client character\_set\_connection character\_set\_database character\_set\_filesystem character\_set\_results character\_set\_server character\_set\_system This is a topic that I don't know much about and I am still far from an expert. My focus has been other DBMS topics. But I spent time recently on this topic while… ## [Sysbench vs MySQL on a small server: no new regressions, many old ones](http://smalldatum.blogspot.com/2026/03/sysbench-vs-mysql-on-small-server-no.html) _2026-03-24 · Mark Callaghan · Small Datum_ This has performance results for InnoDB from MySQL 5.6.51, 5.7.44, 8.0.X, 8.4.8 and 9.7.0 on a small server with sysbench microbenchmarks. The workload here is cached by InnoDB and my focus is on regressions from new CPU overheads. In many cases, MySQL 5.6.51 gets about 1.5X more QPS than modern MySQL (8.0.x thru 9.7). The root cause is new CPU overhead, possibly from code bloat. tl;dr There are… ## [MariaDB innovation: binlog\_storage\_engine, 32-core server, Insert Benchmark](http://smalldatum.blogspot.com/2026/03/mariadb-innovation-binlogstorageengine_0126897374.html) _2026-03-19 · Mark Callaghan · Small Datum_ MariaDB 12.3 has a new feature enabled by the option binlog\_storage\_engine . When enabled it uses InnoDB instead of raw files to store the binlog. A big benefit from this is reducing the number of fsync calls per commit from 2 to 1 because it reduces the number of resource managers from 2 (binlog, InnoDB) to 1 (InnoDB). This work was done by Small Datum LLC and sponsored by the MariaDB Foundation.… ## [MariaDB innovation: binlog\_storage\_engine, 48-core server, Insert Benchmark](http://smalldatum.blogspot.com/2026/03/mariadb-innovation-binlogstorageengine.html) _2026-03-18 · Mark Callaghan · Small Datum_ MariaDB 12.3 has a new feature enabled by the option binlog\_storage\_engine . When enabled it uses InnoDB instead of raw files to store the binlog. A big benefit from this is reducing the number of fsync calls per commit from 2 to 1 because it reduces the number of resource managers from 2 (binlog, InnoDB) to 1 (InnoDB). See this blog post for more details on the new feature. This work was done by… ## [CPU efficiency for MariaDB, MySQL and Postgres on TPROC-C with a small server](http://smalldatum.blogspot.com/2026/03/cpu-efficiency-for-mariadb-mysql-and.html) _2026-03-16 · Mark Callaghan · Small Datum_ I started to use TPROC-C from HammerDB to test MariaDB, MySQL and Postgres and published results for MySQL and Postgres on small and large servers. This post provides more detail on CPU overheads for MariaDB, MySQL and Postgres on a small server. tl;dr Postgres get the most throughput and the difference is large. MariaDB gets more throughput than MySQL Throughput improves for MariaDB and MySQL but… ## [The first rule of database fight club: admit nothing](http://smalldatum.blogspot.com/2026/03/the-first-rule-of-database-fight-club.html) _2026-03-06 · Mark Callaghan · Small Datum_ I am fascinated by tech marketing but would be lousy at it. A common practice is to admit nothing -- my product, project, company, idea is perfect. And I get it because admitting something isn't perfect just provides fodder for marketing done by the other side, and that marketing is often done in bad faith. But it is harder to fix things when you don't acknowledge the problems. I wrote about this… ## [MariaDB innovation: vector index performance](http://smalldatum.blogspot.com/2026/02/mariadb-innovation-vector-index.html) _2026-02-23 · Mark Callaghan · Small Datum_ Last year I shared many posts documenting MariaDB performance for vector search using ann-benchmarks . Performance was great in MariaDB 11 and this blog post explains that it is even better in MariaDB 12. This work was done by Small Datum LLC and sponsored by the MariaDB Foundation. My previous posts were published in January and February 2025. tl;dr Vector search recall vs precision in MariaDB… ## [Explaining why throughput varies for Postgres with a CPU-bound Insert Benchmark](http://smalldatum.blogspot.com/2026/02/explaining-why-throughput-varies-for.html) _2026-02-18 · Mark Callaghan · Small Datum_ Throughput for the write-heavy steps of the Insert Benchmark look like a distorted sine wave with Postgres on CPU-bound workloads but not on IO-bound workloads. For the CPU-bound workloads the chart for max response time at N-second intervals for inserts is flat but for deletes it looks like the distorted sine wave. To see the chart for deletes, scroll down from here . So this looks like a problem… ## [MariaDB innovation: binlog\_storage\_engine, small server, Insert Benchmark](http://smalldatum.blogspot.com/2026/02/mariadb-innovation-binlogstorageengine_17.html) _2026-02-18 · Mark Callaghan · Small Datum_ MariaDB 12.3 has a new feature enabled by the option binlog\_storage\_engine . When enabled it uses InnoDB instead of raw files to store the binlog. A big benefit from this is reducing the number of fsync calls per commit from 2 to 1 because it reduces the number of resource managers from 2 (binlog, InnoDB) to 1 (InnoDB). My previous post had results for sysbench with a small server. This post has… ## [MariaDB innovation: binlog\_storage\_engine](http://smalldatum.blogspot.com/2026/02/mariadb-innovation-binlogstorageengine.html) _2026-02-16 · Mark Callaghan · Small Datum_ MariaDB 12.3 has a new feature enabled by the option binlog\_storage\_engine . When enabled it uses InnoDB instead of raw files to store the binlog. A big benefit from this is reducing the number of fsync calls per commit from 2 to 1 because it reduces the number of resource managers from 2 (binlog, InnoDB) to 1 (InnoDB). In this post I have results for the performance benefit from this when using… ## [HammerDB tproc-c on a large server, Postgres and MySQL](http://smalldatum.blogspot.com/2026/02/hammerdb-tproc-c-on-large-server.html) _2026-02-15 · Mark Callaghan · Small Datum_ This has results for HammerDB tproc-c on a small server using MySQL and Postgres. I am new to HammerDB and still figuring out how to explain and present results so I will keep this simple and just share graphs without explaining the results. The comparison might favor Postgres for the IO-bound workloads because I used smaller buffer pools than normal to avoid OOM. I have to do this because RSS for…