https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Optimizing-reduce-by-key-implementation-using-TBB/m-p/939213#M5041 <P>Hello Everyone,</P> <P>I'm quite new to TBB &amp; have been trying to optimize reduce_by_key implementation using TBB constructs. However serial STL code is always outperforming the TBB code! It would be helpful if I'm given an idea about how reduce_by_key can be improvised using tbb::parallel_scan. Any help at the earliest would be much appreciated.</P> <P>Thanks.</P> Thu, 06 Mar 2014 04:25:34 GMT Shruti_R_ 2014-03-06T04:25:34Z https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Optimizing-reduce-by-key-implementation-using-TBB/m-p/939213#M5041 <P>Hello Everyone,</P> <P>I'm quite new to TBB &amp; have been trying to optimize reduce_by_key implementation using TBB constructs. However serial STL code is always outperforming the TBB code! It would be helpful if I'm given an idea about how reduce_by_key can be improvised using tbb::parallel_scan. Any help at the earliest would be much appreciated.</P> <P>Thanks.</P> Thu, 06 Mar 2014 04:25:34 GMT https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Optimizing-reduce-by-key-implementation-using-TBB/m-p/939213#M5041 Shruti_R_ 2014-03-06T04:25:34Z