I'm using OpenMP with concurrent data structures from Intel TBB. localClustersTable is a concurrent_has_map.
#pragma omp for
for(auto& it = localClustersTable.begin(); it<localClustersTable.end(); it++)
auto& clusters = it->second;
But it doesn't support "<" operator for its iterator. I can not use "!=" instead of "<" because that won't be compatible with OpenMP. So I can't parallelize the loop using OpenMP because OpenMP needs random access to container. How can I make it work?