Or just use a std::vector<double> for sg. That should do it.
Romain
Envoy? de mon iPhone
Le 14 juil. 2015 ? 13:21, JJ Allaire <jj.allaire at gmail.com> a ?crit :
The examples in the RcppParallel documentation assume that access to
vectors and matrixes are *aligned* (i.e. fall into neat buckets
whereby reading and writing doesn't overlap between worker instances).
Your example appears to access arbitrary elements of sg (depending on
what's passed in gi) which probably creates overlapping reads/writes.
You should also study the documentation for join carefully.
There's nothing incorrect about RcppParallel's behavior here, rather
you need to think more carefully about the access patterns of your
data and how they might conflict. You may need to introduce locking to
overcome the conflicts, which in turn could kill the performance
benefit you gain from parallelism. No easy answers here :-\
On Tue, Jul 14, 2015 at 7:15 AM, Danas Zuokas <danas.zuokas at gmail.com>
Yes it is the same question on SO and I did consider RHertel's comments.
But this problem (sums by group id) is not parallelFor it is
I split vector, calculate sums and then aggregate those sums.
Please correct me if I am wrong.
2015-07-14 13:54 GMT+03:00 Dirk Eddelbuettel <edd at debian.org>:
On 14 July 2015 at 09:25, Danas Zuokas wrote:
| I have written parallel implementation of sums in groups using
RcppParallel.
Isn't this the same question as
You got some excellent comments there by SO user 'RHertel'. Did you
consider those?
Dirk
| // [[Rcpp::depends(RcppParallel)]]
| #include <Rcpp.h>
| #include <RcppParallel.h>
| using namespace Rcpp;
| using namespace RcppParallel;
|
| struct SumsG: public Worker
| {
| const RVector<double> v;
| const RVector<int> gi;
|
| RVector<double> sg;
|
| SumsG(const NumericVector v, const IntegerVector gi, NumericVector
sg): v(v), gi(gi), sg(sg) {}
| SumsG(const SumsG& p, Split): v(p.v), gi(p.gi), sg(p.sg) {}
|
| void operator()(std::size_t begin, std::size_t end) {
| for (std::size_t i = begin; i < end; i++) {
| sg[gi[i]] += v[i];
| }
| }
|
| void join(const SumsG& p) {
| for(std::size_t i = 0; i < sg.length(); i++) {
| sg[i] += p.sg[i];
| }
| }
| };
|
| // [[Rcpp::export]]
| List sumsingroups(NumericVector v, IntegerVector gi, int ni) {
| NumericVector sg(ni);
| SumsG p(v, gi, sg);
| parallelReduce(0, v.length(), p);
|
| return List::create(_["sg"] = p.sg);
| }
|
| It compiles using Rcpp::sourceCpp. Now when I call it from R
sumsingroups(1:10,
| rep(0:1, each = 5), 2) several times I get the right answer (15 40)
then
| something different (usually some multiplicative of the right
Running
|
|
| res <- sumsingroups(1:10, rep(0:1, each = 5), 2)
| for(i in 1:1000) {
| tmp <- sumsingroups(1:10, rep(0:1, each = 5), 2)
| if(res[[1]][1] != tmp[[1]][1]) break
| Sys.sleep(0.1)
| }
|
| breaks at random iteration returning
|
| $sg
| [1] 60 160
|
| or
|
| $sg
| [1] 30 80
|
| I am new to Rcpp and RcppParallel and do not know what could cause
| behavior.
|
| Things that did not help:
|
| 1. Added for (std::size_t i = 0; i < sg.length(); i++) sg[i] = 0; to
both of
| constructors.
| 2. Changed names so that they are different in Worker definition
| function implementation.
|
| _______________________________________________
| Rcpp-devel mailing list
| Rcpp-devel at lists.r-forge.r-project.org
|