The behavior of vm.swappiness was always a bit confusing for novice linux users, as setting vm.swappiness to 0 would not completely disable swapping in the system during a memory crunch. vm.swappiness would only affect the agressiveness of swapping.
Following upstream commit tried to give more control to parameter. This commit tried to avoid swapping completely when vm.swappiness is set to 0.
Having done my mini-series on Nested Loop join logical I/O optimizations a while ago I unfortunately never managed to publish anything regarding the Nested Loop join physical I/O optimizations, which are certainly much more relevant to real-life performance.Therefore the main purpose of this blog post is to point you to Nikolay Savvinov's (whose blog I can recommend in general) great mini-series covering various aspects of these optimizations:Part 1
style="font-weight: 400;">The breaking down of monolithic code bases into rapidly deployable microservices marks a paradigm shift in software development. As described in the first post of this series, microservices allow for faster release cycles and can help squeeze more performance from your hardware footprint. Realizing their full potential, however, will require you to also introduce Docker containers into the mix.
The first half of this year was a little quiet for me on the presenting front. I was concentrating on writing and also on organising events, as opposed to going to them, so most of my trips were for personal reasons (that means “holidays”…). I presented at the Ireland conference and a few UK user group events but that was it – quite a few European events this spring fell on dates I was not available (including the Israeli and Finnish conferences where I was asked to attend and would have loved to). Or, oh the shock of it, my submissions were not accepted!
A simple command like 'DROP TABLESPACE TEMP;' in a regular database can become excessively slow. The reason behind this is quite simple, even if the users defaults to a different temporary tablespace there could still be sessions holding temporary segments there. It's just a matter of finding them and killing them to free the blocks on the temporary segments and you're good to go.
The operation of dropping the temporary tablespace looks like this on Enterprise Manager:
This happened out of the blue after a switchover back from a Datagurad environment. Basically this is due to a bug, the patch is already included in 184.108.40.206.0 and 220.127.116.11.0 patch sets, but it is not enabled by default, it is activated after setting a parameter in the spfile (pfile) parameter file.