News for hackers. Stuff that matters.

A comment on The benefits and costs of writing a POSIX kernel in a high-level language. Parent comment

Re:The benefits and costs of writing a POSIX kernel in a hig... by gmueckl on Monday October 08, 2018 @10:13PM (#18172850)
Sometimes it is hard to understand what a relative 1% can be in absolute terms. Imagine Amazon or Google running 10% slower software in their cloud stacks. That would translate immediately to one or more additional data centers that they would need to build and run and pay for in order to compensate for that.
[ Parent | on HN ]
Re:The benefits and costs of writing a POSIX kernel in a hig... by weberc2 on Monday October 08, 2018 @11:59PM (#18173230)
True, but 10% slower OS is not a 10% slower program. Most time spent is not likely in the kernel.
[ Parent | on HN ]
Re:The benefits and costs of writing a POSIX kernel in a hig... by gmueckl on Tuesday October 09, 2018 @12:05AM (#18173254)
If you make this argument about any single component, you end up with low performance across the whole system because it leads to a culture where everybody can point out that their code uses only a few percent of the CPU.
[ Parent | on HN ]
Re:The benefits and costs of writing a POSIX kernel in a hig... by MaxBarraclough on Tuesday October 09, 2018 @05:25AM (#18174428)
A valid point, as we should be frustrated that (in desktop computers, say) real user-facing performance hasn't been keeping pace with improvements in hardware.

Every decade, hardware performance skyrockets upward, but developers adopt libraries/frameworks/languages/low standards/habits which give rise to woefully inefficient software. It's so atrocious that it seems to roughly counterbalance the hardware advances. Heavyweight single-page web applications seem to be the endgame of this trend, for now.

See Wirth's law: What Intel giveth, Microsoft taketh away.

That said, the 80-20 rule is a real thing. It's possible that you'll get better real performance by writing 80% of your code in Python and the performance-critical 20% in carefully optimised C, than by spending the same amount of time writing it all in Java.

[ Parent | on HN ]
Re:The benefits and costs of writing a POSIX kernel in a hig... by weberc2 on Tuesday October 09, 2018 @08:07AM (#18175214)
Sure. I just wanted to clarify that 10% OS perf != 10% application perf.
[ Parent | on HN ]
Re:The benefits and costs of writing a POSIX kernel in a hig... by davrosthedalek on Tuesday October 09, 2018 @10:34AM (#18176276)
On the other hand, a 10% faster OS makes /every/ program faster.
[ Parent | on HN ]