Forest @forestjohnson@pixie.town
- member of
- https://cyberia.club
- webmaster @
- https://capsul.org/
I am a web technologist who is interested in supporting and building enjoyable ways for individuals, organizations, and communities to set up and maintain their own server infrastructure, including the hardware part.
I am currently working full time as an SRE 😫, but I am also heavily involved with Cyberia Computer Club and Layer Zero
Joined Feb 2021
@f0x are you working on somethting for gotosocial?
I did some benchmarks on my "for fun" spatial database index today: https://sequentialread.com/building-a-spatial-index-supporting-range-query-using-space-filling-hilbert-curve/#benchmarks
@dumpsterqueer This is cool!!
my techbro shit reaction:
what's the point of TLS if some CAs/DNS operators somewhere own all the keys to kick folks off the network, impersonate them, etc?
Obviously cloudflare is worse because its impossible tell how much they are processing/recording your traffic, while bona-fide attacks against TLS/x.509 are harder to hide..
at any rate, trying to make my own mini cloudflare service where the user exclusively owns the keys, and maybe has automated monitoring to spot the x.509 authority attacks if/when they happen 😬
more social media demo app dev stream today https://stream.sequentialread.com/
this is my kid, millisecondsSinceUnixEpoch. He's a real straight-forward lad
any.... space filling curvers in the chat ???
@f0x It might be illuminating to collect a metric for "cache misses caused by previous cache evictions". You would just have to store the cache keys for longer than the actual cache entries, or something.
@f0x 90% load reduction sounds a lot cooler than 26.4% load reduction. Gotta look on the bright side lol 😀
@f0x ok I think get it, so there are two different kinds of cache-misses, the ones where it has to talk to your synapse (content that was posted on your server) aka "synapse" in your grafana dashboard.
Then there are the ones where it has to talk to other matrix servers (i'm assuming those are the ones labeled "remote" in your grafana dashboard).
So it looks like most of the time it's talking to remote servers, not yours -- "synapse" is only like 10% of requests. Isn't that a big win then? 90% bandwidth load reduction for the synapse server?
@f0x FWIW this question is based on an almost-zero understanding of how matrix works lol.
@f0x Do you think the cache hit rate is low because the majority of requests are for never-before-seen content? or because of something else ?
Does the proxy shield your synapse from having to go and fetch thumbnails or images from other servers? i.e. the proxy "federates" with the other servers directly?
I already read the ReadMe and I was slightly confused, I think adding a diagram or definitions of the terms you are using would help.
This is what happens when I let myself work on whatever I want for fun:
Better Portable Graphics (BPG) on the web with WebAssembly (WASM) and ServiceWorker
Music today: 🪣🍗 Buckethead 🎸🌌 Stop by and say hi!
https://stream.sequentialread.com/
stream continues today, more web app dev https://stream.sequentialread.com/
@gabek thats an unpopular opinion ? I always thought that ppl "tolerated" irc the same way everyone tolerates email -- its obviously not ideal but the friction is so great that ppl just hold their nose -- I just assumed that any sort of "extraordinary" needs, like for privacy, resilience, ownership, etc were being addressed by simply looking outside of the "default" irc server. seems reasonable to me
- member of
- https://cyberia.club
- webmaster @
- https://capsul.org/
I am a web technologist who is interested in supporting and building enjoyable ways for individuals, organizations, and communities to set up and maintain their own server infrastructure, including the hardware part.
I am currently working full time as an SRE 😫, but I am also heavily involved with Cyberia Computer Club and Layer Zero
Joined Feb 2021
