226 IBM eX5 Implementation Guideprocessing nodes in a high-performance cluster, database servers, or print servers forgraphics printers require the best memory performance possible.Remember these considerations when installing memory in the System x3850 X5: Due to the fact that no single processor has direct access to all PCIe slots, the QPI linksthat are used to communicate between the two processors are not only used for memorycalls, but also for interfacing to PCIe adapters. When installing memory for multiple processors and not all memory cards are installed,only processing threads that are assigned to processor without local memory experiencea 50% increase in memory latency. For a server with heavy I/O processing, the additionalmemory traffic also affects the efficiency of addressing PCIe adapters. Any memory that is installed on memory cards for an uninstalled or defective processor isnot seen by the other processors on the server. For nonuniform memory access (NUMA)-aware operating systems when multipleprocessors are installed, you must install the same amount of memory on each memorycard of each installed processor. See 3.8.2, “DIMM population sequence” on page 79 fordetails. You can achieve the best processor performance when memory is installed to supportHemisphere Mode. Hemisphere Mode is the required memory configuration to allow two3850 X5 servers to scale. It is possible to have processor 1 and 4 in Hemisphere Mode,and not processor 2 and 3, to permit scaling. In this type of installation, having processor 1and 4 in Hemisphere Mode improves the memory access latency for all processors. Todetermine the DIMM population for Hemisphere Mode, see 3.8.2, “DIMM populationsequence” on page 79. You might consider installing all DIMMs of the same field-replaceable unit (FRU) to avoidconflicts that might prevent NUMA compliance or Hemisphere Mode support. Theseproblems are most likely to occur when multiple DIMM sizes are installed. The serversupports installing multiple-sized DIMMs in each rank, but the configuration becomescomplex and difficult to maintain. Operating systems that depend on NUMA complianceinform you when the server is not NUMA-compliant. However, nothing informs you that theprocessors are not in Hemisphere Mode. It is better to install more smaller DIMMs than fewer larger DIMMs to ensure that all of thememory channels and buffers of each processor have access to the same amount ofmemory. This approach allows the processors to fully use their interleave algorithms toaccess memory faster by spreading access over multiple paths.6.3.1 Testing the memory DIMMsThe DIMM components on a server are the most sensitive to static discharge. Improperhandling of memory is the most likely cause of DIMM failures. Working in an extremely drylocation dramatically increases the possibility that you will build up a static charge. Alwaysuse an electrostatic discharge (ESD) strap connected between you and a grounding point toreduce static buildup.The best practice when installing memory is to run a memory quick test in diagnostics toensure that all of the memory is functional. The following reasons describe why memorymight not be functional: Wrong DIMM for the type of server that you have. Ensure that only IBM-approved DIMMsare installed in your server. The DIMM is not fully installed. Ensure that the DIMM clips are in the locked position toprevent the DIMM from pulling out of its slot.
