TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

ZGC: A Scalable Low-Latency Garbage Collector [video]

67 点作者 javinpaul超过 6 年前

4 条评论

bradleyjg超过 6 年前
The talk has a lot of interesting technical details, but the bottom line for java developers or jvm sys admins is that the new GC trades off throughput for reduced maximum pause times (current goal is 10ms max pause time, longer term goal is less than 1 ms).<p>The upshot of that is that you <i>probably</i> don&#x27;t want to swap to this unless you use a relatively large heap (at least 32 GB, maybe more), because below that level the STW latency of the other GC options aren&#x27;t <i>generally speaking</i> bad enough to be worth the trade-off.<p>Also if latency is a big concern for you and you&#x27;ve got dollars to spend consider Azul zing as well. (That part isn’t from the talk.)
评论 #18323049 未加载
nickpsecurity超过 6 年前
Slides for folks that don&#x27;t want to watch a video:<p><a href="https:&#x2F;&#x2F;cr.openjdk.java.net&#x2F;~pliden&#x2F;slides&#x2F;ZGC-Jfokus-2018.pdf" rel="nofollow">https:&#x2F;&#x2F;cr.openjdk.java.net&#x2F;~pliden&#x2F;slides&#x2F;ZGC-Jfokus-2018.p...</a>
rayiner超过 6 年前
It&#x27;s interesting that the conditional read barrier is so cheap (4% overhead on SPECjbb). In past work, read barriers have been considered relatively high overhead and thus algorithms dependent on them avoided for that reason. For example: <a href="https:&#x2F;&#x2F;hosking.github.io&#x2F;links&#x2F;Yang+2012ISMM.pdf" rel="nofollow">https:&#x2F;&#x2F;hosking.github.io&#x2F;links&#x2F;Yang+2012ISMM.pdf</a> (see Table 3 on page 41).
评论 #18422859 未加载
评论 #18322591 未加载
评论 #18323017 未加载
geodel超过 6 年前
I expect Rust&#x27;s resident GC expert enlighten people working on ZGC why their approach is wrong in trading throughput for low latency. And that they need to learn from hotspot GC.
评论 #18323616 未加载
评论 #18321463 未加载
评论 #18323392 未加载
评论 #18323091 未加载
评论 #18321987 未加载