Go GC: Prioritizing low latency and simplicity

Re:Go GC: Prioritizing low latency and simplicity by mholt on Tuesday September 01, 2015 @12:14AM (#10150689)

I'm no academic (just a CS student), but from the very little I understand about GC algorithms, there's no free lunch. In other words, choosing GC isn't like shopping for a new condo... different algorithms have different advantages and implementation difficulties. There may also be caveats with some algorithms that others aren't plagued by. "Better" means different things to different people and in different situations.

For what it's worth, many of today's finest, most important algorithms are from the 50s-70s, or maybe special adaptations of them. Theory doesn't obsolete like technologies do.

No need to get depressed by this. Hard problems are still hard (until proven otherwise), and taking out the trash is the same today as it has ever been.

[ Parent | on HN ]

Re:Go GC: Prioritizing low latency and simplicity by eternalban on Tuesday September 01, 2015 @09:28AM (#10152135)

> For what it's worth, many of today's finest, most important algorithms are from the 50s-70s, or maybe special adaptations of them.

This is only the case/sensible for algorithms that have not been practically improved.

[ Parent | on HN ]