I was quite happy with my performance, give it a listen/view:
I’m currently working on a Brillo audio component, which will require very low latency performance. While searching for some tips and prior art, I came across this talk from Google I/O 2013. A bit older now, but provides an excellent overview of the problem domain and approaches to solving those issues..
This is a pretty entertaining talk from Chandler Carruth at CppCon 2014 - lots of good takeaways..
This talk is incredibly informative - Twitter have their own fork of OpenJDK, in which they have enabled registers on the CPU, normally used by the Java Compiler - this enables Frame Pointers which perf can read and translate, enabling a full Stack trace from JVM bytecode right down into the kernel.
Beyond simply CPU counters and stack trace, they also tie in other JVM flags which export DTrace counters, and use these to construct connections between memory allocation and the running process, so in the end you have a tool which can spans JVM -> kernel connections, alongside CPU -> memory.
Sounds very useful, i look forward to them open-sourcing it..
The always interesting Brendan Greg —
I was along at this talk last year, and just now found it online - was one of the most informative talks i've been to, learned loads from it -
Been working a lot of JVM GC performance problems of late, so this video interview my friend Johan turned me on has some great info -
Just got back from the Usenix/Lisa 12 conference in San Diego, and had a great time, super inspiring talks and content.
Highlight of the conference for me was Brendan Gregg speaking on Performance Analysis Methodologies - most of his talk was based upon a paper he just published in ACM - Thinking Methodically About Performance.
The talks haven't yet been published on the Usenix website, but Brendan's blog has a ton of great looking content and older talks including this one on Visualisations for Performance Analysis