Hi Loïc,totaly agree about performances. honestly, we don't experiment yet any trouble in our erasure coding implementation (which is just software optimized for the moment). the bottleneck is more located at the network (basically at 1Gbit/s) and at the disks i/o levels. and, I guess it is the same conclusion for well optimized libraries like Jerasure. to continue about Mojette transform, this erasure code is in fact (1+epsilon)MDS with a linear complexity both in coding and decoding (we just need addition and subtraction in any field). furthermore, it is well implemented in RozoFS. th epsilon (the overhead) is very negligible for normal use cases.
best, bp Le 16/06/2013 23:31, Loic Dachary a écrit :
Hi Benoît, From the ( naïve ) point of view of engineering, performances are important. The recent works of James Plank ( cc'ed ) greatly improved them and I'm looking forward to the next version of jerasure ( http://web.eecs.utk.edu/~plank/plank/papers/CS-08-627.html ). Rozofs Mojette Transform implementation ( https://github.com/rozofs/rozofs/blob/master/rozofs/common/transform.h & https://github.com/rozofs/rozofs/blob/master/rozofs/common/transform.cc ) does not seem to make use of SIMD. Is it because the performances are good enough to not require them ? Cheers On 06/16/2013 09:51 PM, Benoît Parrein wrote:Paul Von-Stamwitz <PVonStamwitz <at> us.fujitsu.com> writes: Hi Paul,Loic, I know nothing about Mojette Transforms. From what little I gleaned,it might be good for repair(needing only a subset of chunks within a range to recalculate a missingchunk) but I'm worried about thestorage efficiency. RozoFS claims 1.5x. I'd like to do better than that. All the best, PaulIf you want to do better than that you will probably lose in availability. 1.5x give the same availability than 3 replicats and that for any kind of erasure coding. FYI, Mojette transform has no constraint in terms of Galois fields. It is the big advantage to use discrete geometry rather than algebra. best regards, bp -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
<<attachment: benoit_parrein.vcf>>
