Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.programming > #4605 > unrolled thread
| Started by | Ramine <ramine@1.1> |
|---|---|
| First post | 2014-06-14 13:06 -0700 |
| Last post | 2014-06-14 14:28 -0700 |
| Articles | 2 — 1 participant |
Back to article view | Back to comp.programming
About RCU... Ramine <ramine@1.1> - 2014-06-14 13:06 -0700
Re: About RCU... Ramine <ramine@1.1> - 2014-06-14 14:28 -0700
| From | Ramine <ramine@1.1> |
|---|---|
| Date | 2014-06-14 13:06 -0700 |
| Subject | About RCU... |
| Message-ID | <lni9vl$cis$1@dont-email.me> |
Chris Thomasson wrote: >IMVHO, you do not seem understand why RCU beats a non-asymmetric >rw-lock. Besides that, how can you compare RCU with a read-write mutex >when RCU allows reads and writes to occur concurrently? >Can a write proceed when a read is in progress in any of your rw-mutex >algos? That's not true in general, RCU uses some technics that permit parallelism like a write proceed when reads are in progrogress, but the RCU technic don't apply to the general cases, if you are writing and reading to multiple memory addresses and they are conflicts, that means you are writing to many same memory addresses so RCU will not run a write and reads in parallel... And i think on read mostly scenarios the difference between RCU and my scalable RWLocks are small, cause my scalable RWLocks are scalable on read mostly scenarios , they scale even at 3% of writes. Thank you, Amine Moulay Ramdane.
[toc] | [next] | [standalone]
| From | Ramine <ramine@1.1> |
|---|---|
| Date | 2014-06-14 14:28 -0700 |
| Message-ID | <lnieq0$djq$1@dont-email.me> |
| In reply to | #4605 |
On 6/14/2014 1:06 PM, Ramine wrote: > > Chris Thomasson wrote: > >IMVHO, you do not seem understand why RCU beats a non-asymmetric > >rw-lock. Besides that, how can you compare RCU with a read-write mutex > >when RCU allows reads and writes to occur concurrently? > > >Can a write proceed when a read is in progress in any of your rw-mutex > >algos? > > > That's not true in general, RCU uses some technics that permit > parallelism like a write proceed when reads are in progrogress, but > the RCU technic don't apply to the general cases, if you are > writing and reading to multiple memory addresses and they are conflicts, > that means you are writing to many same memory addresses so > RCU will not run a write and reads in parallel... > > And i think on read mostly scenarios the difference between RCU and my I mean the difference on performance of course. > scalable RWLocks are small, cause my scalable RWLocks are scalable on > read mostly scenarios , they scale even at 3% of writes. > > > > Thank you, > Amine Moulay Ramdane.
[toc] | [prev] | [standalone]
Back to top | Article view | comp.programming
csiph-web