Hacker News new | past | comments | ask | show | jobs | submit login

Anything around scaling across multiple cores I suppose. It wasn't as nuanced back then.

In 2004 single core computers were still quite prevalent, and the idea of steady significant performance improvements coming from increasing clock speeds was in its death throes. From memory it was about then that Intel had given up on the P4 Netburst architecture and stepped back to improving the P3 to release the new Core and Core Duo architectures.

Nowadays multicore has been reality for a long time and the number of cores in a CPU has increased a lot, and we're thinking much more about different competing ways of achieving better concurrency. Back then it was just about getting better multicore concurrency at all using bog standard threads.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: