GopherCon 2015: Go GC: Solving the Latency Problem - Rick Hudson
Why Hacker News recommends it
Pointed to as the authoritative source on Go's GC pause improvements, and later revisited when GC latency claims were challenged.
What it is
Rick Hudson's GopherCon 2015 talk on Go 1.5's new low-latency concurrent garbage collector. It covers the motivation, performance numbers, technical challenges and why the approach fits Go.
Recommended by 2 HN commenters across 2 different months, 2015 to 2020.
What Hacker News said
Recommends shawnps · 2015-10 · on “Golang proposal: Dense mark bits and sweep-free allocation”
If you liked this you might also find Rick Hudson's talk on GC improvements in Go 1.5 interesting: [this video] [LINK]
Recommends kelseyhightower · 2016-01 · on “Google Open Source Load Balancer in Go”
Go's TLS support should be greatly improved in Go 1.6 thanks to the performance work you referenced by Cloudflare. The licensing issue was resolved and the work was merged 11/10/2016[1] While GC will always carry a cost, great improvements have been made over the last year and continues to be an area of focus for each Go release. Rick Hudson gave a nice update on Go's GC improves at GopherCon 2015 [2][3] [1] [LINK] [2] [LINK] . [3] [this video]
Cites twotwotwo · 2015-08 · on “Go 1.6 GC improvements”
It looks like a few things in there could further reduce typical and peak pause times, by backgrounding more phases of the work and rescanning stacks late so it's less likely much marking has to happen in the pause. I'm less clear on this, but "embrace write barrier during scan" and the other pacing changes might also better spread out work during the concurrent phases. The pause improvements in 1.5 are no joke -- here's a benchmark from the GopherCon talk on it: [LINK] (1.5 is the yellow dots along the bottom) (That talk is [this video] and the slides are [LINK] ) And here's a graph from a real-world memory hungry app on a beta--300ms pause down to 4ms: [LINK] (Presentation on that app here: [LI
Cites gizzlon · 2016-05 · on “The Path to Rust”
Oki, got it. Can't decide if that sounds like horrible conditions or a fun challenge =) For reference, the Go 1.5 GC propaganda: > Go 1.5, the first glimpse of this future, achieves GC latencies well below the 10 millisecond goal we set a year ago. Blog post: [LINK] Slide: [LINK] Talk: [this video] Edit: Meanwhile, over in Java land they measure stop the world pauses in seconds , not milliseconds: [LINK] [LINK]
Critical deepsun · 2020-02 · on “Why Discord is switching from Go to Rust”
Wait, isn't Go devs said they solved GC latency problems [1]? (from 2015): "Go is building a garbage collector (GC) not only for 2015 but for 2025 and beyond: A GC that supports today’s software development and scales along with new software and hardware throughout the next decade. Such a future has no place for stop-the-world GC pauses, which have been an impediment to broader uses of safe and secure languages such as Go." [2] [1] [this video] [2] [LINK]