A JVM Does That? (2011) [pdf]

Re:A JVM Does That? (2011) [pdf] by cheez on Thursday July 07, 2016 @05:56PM (#12052323)

> I had one CTO of a public HFT firm tell me that short of FPGA's Azul's pause-less GC (https://www.azul.com/products/zing/pgc/) was the biggest performance win they'd had in the past few years, note: this was a few years ago.

Even better than using C++? Something sounds off.

[ Parent | on HN ]

Re:A JVM Does That? (2011) [pdf] by chollida1 on Thursday July 07, 2016 @07:47PM (#12052793)

I don't know where you got C++ from.

In any case, I assume they stayed with Java as their entire code base was Java with the exception of what ever they program their FPGAs with.

[ Parent | on HN ]

Re:A JVM Does That? (2011) [pdf] by TickleSteve on Friday July 08, 2016 @01:52AM (#12053770)

FPGA's are typically used for the real-time front-end in systems like this....

...Which means that the Java side wasn't doing anything particularly quick or real-time.

The Java side was the 'easy&slow' side.... hence they could use Java.

[ Parent | on HN ]

Re:A JVM Does That? (2011) [pdf] by cheez on Friday July 08, 2016 @05:19AM (#12054372)

Yeah, I concur with what the other guy says. There is no way they were using Java for the HFT side.

[ Parent | on HN ]

Re:A JVM Does That? (2011) [pdf] by duaneb on Thursday July 07, 2016 @06:09PM (#12052390)

C++ doesn't improve on I/O (including cache misses).

[ Parent | on HN ]

Re:A JVM Does That? (2011) [pdf] by gpderetta on Thursday July 07, 2016 @06:14PM (#12052407)

HFT systems aren't normally IO bound.

[ Parent | on HN ]