https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Memory-access-pattern-for-threads/m-p/1018155#M6604 <P>On typical numa platforms, including Xeon multi socket since introduction of qpi, additional cycles are spent on remote access.</P> Thu, 28 May 2015 13:30:55 GMT TimP 2015-05-28T13:30:55Z https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Memory-access-pattern-for-threads/m-p/1018154#M6603 <P><SPAN style="font-size: 1em; line-height: 1.5;">&nbsp; &nbsp; &nbsp; I've one basic question on threads memory access pattern. Suppose the computer/system/node has two sockets, each socket has its own block of memory(shared among two sockets), each socket has 4 cores. If there are two threads running(forked from a single process, may be pthreads/openmp threads), and thread 1 is on socket 1 and thread 2 on socket 2. If thread 1 tries to access data from socket 2's block of memory, then whether access time for this is same as accessing the data from its own block of memory or different?&nbsp;</SPAN></P> Thu, 28 May 2015 10:52:50 GMT https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Memory-access-pattern-for-threads/m-p/1018154#M6603 Sangamesh_B_ 2015-05-28T10:52:50Z https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Memory-access-pattern-for-threads/m-p/1018155#M6604 <P>On typical numa platforms, including Xeon multi socket since introduction of qpi, additional cycles are spent on remote access.</P> Thu, 28 May 2015 13:30:55 GMT https://community.intel.com/t5/Intel-Moderncode-for-Parallel/Memory-access-pattern-for-threads/m-p/1018155#M6604 TimP 2015-05-28T13:30:55Z