Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.programming > #3864 > unrolled thread
| Started by | aminer <aminer@toto.net> |
|---|---|
| First post | 2013-10-04 21:24 -0700 |
| Last post | 2013-10-04 21:24 -0700 |
| Articles | 1 — 1 participant |
Back to article view | Back to comp.programming
About scalability and Locks... aminer <aminer@toto.net> - 2013-10-04 21:24 -0700
| From | aminer <aminer@toto.net> |
|---|---|
| Date | 2013-10-04 21:24 -0700 |
| Subject | About scalability and Locks... |
| Message-ID | <l2npmu$ae8$1@news.albasani.net> |
Hello again, I think i have implemented all types of Locks: Spinlocks with a backoff mechanism, the array based locks, and distributed locks etc. and i think i have understood them well: the fair Locks such as the Ticket spinlock and the Anderson Lock and the MCS lock and my scalable distributed Lock that is fair when there is contention and that avoid starvation all of them do not scale and become slow if the number of threads are greater than the number of cores, but the unfair locks such as the Spinlock with backoff and my scalable distributed Lock(the version that is unfair) will scale even if the number of threads are greater than the number of cores. This is very important to understand. I have implemented an enhanced version of the scalable Anderson Lock and i will put it tomorrow on my website. So be smart please... Thank you, Amine Moulay Ramdane.
Back to top | Article view | comp.programming
csiph-web