mnutt

11 ms·

91.
by mnutt 4y ago
Nothing but lots of small (~100b) pipelined SETs and a small number of GETs here and there. Only 10MB/s but at 100k SETs/sec redis’s CPU core sits at 60-70%. Active defrag can easily send it into a death spiral.
92.
by mnutt 4y ago
There are workloads that will saturate a redis instance's CPU: using it as an LRU cache, eventually you will hit the configured memory limits and adding new keys will require finding old keys to delete. Eventually it may also require r
93.
by mnutt 4y ago
A few years ago I worked on a web-based augmented reality app intended to be shared by users through services that often displayed links in webviews. Newer SFSafariViewController worked fine but WKWebView failed silently in a way that was i
94.
by mnutt 4y ago
Sandstorm does something similar in its packaging process. In practice it works pretty well; besides integration tests exercising functionality, the packaging generates a manifest list people typically check into source control, making it e
95.
by mnutt 4y ago
A long time back I entertained the idea of buying hc.im and using the email address le@hc.im. (my name is michael) Perhaps making business cards with a laser cutout so I could turn the card over if it wasn't apparent. But two-letter im
96.
by mnutt 4y ago
Are you sure? My understanding was that nginx would fill up all free connections up to the max, but then would begin draining idle connections so it could create new ones for new visitors.
97.
by mnutt 4y ago
Multi-process in node.js web serving is doable but operationally more of a pain. The cluster module will let you spawn a bunch of processes all sharing the same listening port, but sharing caches, connection pools, etc become much more diff
98.
by mnutt 5y ago
You can even automate this with something like Percy that takes screenshots and does visual diffs at selected points in the test suite. You just have to be careful about including random/date-based data in your tests or you’ll get more
99.
by mnutt 5y ago
Do you have any sense of the expected performance of web streams as compared to the existing node.js streams?
100.
by mnutt 5y ago
Many of the Design Sprint brainstorming sessions I've been a part of felt superficial. I can see the value in pushing people to list out lots of ideas very quickly, to force them to explore the solution space and not get stuck on a sin
101.
by mnutt 5y ago
I wish GitHub would provide some sort of recommended projects based on star clustering. They have something involving stars on the GitHub Explore page, but the results are not very good.
102.
by mnutt 5y ago
My experience has been that using canvas to resize images produces inconsistent results from one browser to another, and in most cases is higher file size and lower quality than other tools. Enabling imageSmoothingEnabled helps, but is only
103.
by mnutt 5y ago
QtWebKit is independently maintained now and intends to support Qt6 in the future.
104.
by mnutt 5y ago
There's an argument to be made for a tradeoff between legacy device support and cleaning up cruft, but the part that stands out to me is that optional chaining is such a trivial thing to break backwards compatibility over. It's on
105.
by mnutt 5y ago
Yes, that was it.
106.
by mnutt 5y ago
There was a startup I talked to at one point that had a service where they’d run an agent on your instances that would collect performance data and live tune kernel parameters, and they had some AI to find the best parameters for your workl
107.
by mnutt 5y ago
That would work, too. It would be very expensive at scale and would be difficult to keep tail latencies down. Which might be ok for some cases, but wasn't for mine.
108.
by mnutt 5y ago
If you need random access to s3 within an ec2 region, sticking nginx in front of it as a caching proxy is unbelievably faster than hitting s3 directly; even moreso if you're multi-region. I recently experimented with trying to have ngi
109.
by mnutt 5y ago
First off, this is really cool. I have been hoping for something like this for a while. Given that the Isolate model has already been developed, have you considered a hybrid isolate/process model, where workers created by a single user
110.
by mnutt 5y ago
My recollection is that Cloudflare Workers uses Isolates, but also employs various process-based strategies. More detail here: https://developers.cloudflare.com/workers/learning/security-...
111.
by mnutt 5y ago
Cloudflare Workers have their own additional compelling value prop in operating on the edge. But I think something like this could offer a lot of value for a different use case: In building complex SaaS apps, there's usually a constant
112.
by mnutt 5y ago
Using CloudShell / the AWS Console for anything involved sounds painful though, among other reasons due to the 12-hour console session timeout. I suppose you might be able to use screen to mitigate it, but getting logged out right in t
113.
by mnutt 5y ago
I believe there are hypervisor.framework backends for vagrant, but it seems that for whatever reason they haven’t taken off? I’m not exactly sure why they haven’t become the default on Mac, given how much more efficient it presumably is.
114.
by mnutt 5y ago
I thought it was interesting how Spotify acquihired uTorrent's creator, Ludvig Strigeus, to build the Spotify desktop client. uTorrent was tiny and fast and written in c++, and early versions of Spotify were also small and incredibly s
115.
by mnutt 5y ago
Varnish has defaulted to streaming responses since varnish 4. I think it gets used for a lot of video streaming use cases.
116.
by mnutt 5y ago
In varnish, if you have some requirements flexibility you can enable grace mode in order to serve stale responses but update from the origin, and avoid long requests every [5] seconds. Not quite the same layer, but in node.js I’m a fan of t
117.
by mnutt 5y ago
My understanding is that mapbox-gl-native dropped node binding support prior to the fork; the code is still there but the build system can’t build it. I’m hoping to help maplibre bring it back up to a working state.
118.
by mnutt 5y ago
I see some analogues in Sandstorm.io, for the web: they’re self-hosted web apps, where instead of the app being the atomic unit, each document (“grain”) is separately managed by sandstorm, and there’s a separate instance of the app running
119.
by mnutt 5y ago
While I see a lot of comparisons to htop, I wonder how this compares to sysdig? I’ve found sysdig to be an invaluable swiss army knife of debugging performance issues; the only difficult part is finding the right documentation among the clo
120.
by mnutt 5y ago
I recently came across a 15 year old bug in qt involving both time comparison _and_ cache invalidation: the code was trying to set timers to fire when the next cache entry was set to expire, but the timeout was in seconds while the eviction

More ›