IBM 71413SU - System x3950 M2 Reference

Contents

24 IBM System x3950 M2 and x3850 M2 Technical IntroductionThere are a number of advanced features implemented in the x3850 M2 and x3950 M2memory subsystem, collectively known as Active Memory: Memory ProteXionThe Memory ProteXion feature (also known as redundant bit steering) provides theequivalent of a hot-spare drive in a RAID array. It is based in the memory controller, and itenables the server to sense when a chip on a DIMM has failed and to route the dataaround the failed chip.Normally, 128 bits out of every 144 are used for data and the remaining 16 bits are usedfor ECC functions. However, the x3850 M2 and x3950 M2 needs only 12 bits to performthe same ECC functions, thus leaving four bits free. In the event that a chip failure on theDIMM is detected by memory scrubbing, the memory controller can reroute data aroundthat failed chip through these spare bits.It can do this automatically without issuing a Predictive Failure Analysis® (PFA) or lightpath diagnostics alert to the administrator, although an event is logged to the serviceprocessor log. After the second DIMM failure, PFA and light path diagnostics alerts wouldoccur on that DIMM as normal. Memory scrubbingMemory scrubbing is an automatic daily test of all system memory, which detects andreports memory errors that might be developing before they cause a server outage.Memory scrubbing and Memory ProteXion work in conjunction with each other and do notrequire memory mirroring to be enabled in order to work properly.When a bit error is detected, memory scrubbing determines if the error is recoverable ornot.– If the error is recoverable, Memory ProteXion is enabled and the data that was storedin the damaged locations is rewritten to a new location. The error is then reported sothat preventative maintenance can be performed. As long as there are enough goodlocations to allow the proper operation of the server, no further action is taken otherthan recording the error in the error logs.– If the error is not recoverable, then memory scrubbing sends an error message to thelight path diagnostics, which then turns on the proper lights and LEDs to guide you tothe damaged DIMM. If memory mirroring is enabled, then the mirrored copy of the datafrom the damaged DIMM is used until the system is powered down and the DIMM isreplaced.Because x3850 M2 and x3950 M2 are now capable of supporting large amounts ofmemory, IBM has added the Initialization Scrub Control setting to the BIOS, to letcustomers choose when this scrubbing is done and therefore potentially speed up theboot process. Refer to Table 10 on page 26 for further detail about these settings. Memory mirroringMemory mirroring is roughly equivalent to RAID-1 in disk arrays, in that usable memory ishalved and a second copy of data is written to the other half. If 8 GB is installed, then theoperating system sees 4 GB after memory mirroring is enabled. It is disabled in the BIOSby default. Because all mirroring activities are handled by the hardware, memory mirroringis operating system-independent.When memory mirroring is enabled, certain restrictions exist with respect to placementand size of memory DIMMs and the placement and removal of memory cards. Refer toSystem x3950M2 and x3850 M2 Installation Guide for details.