Concept Guide
13 Memory Errors and Dell PowerEdge YX4X Server Memory RAS Features
Memory channels must be populated with all one DIMM or all two DIMMs (for example, 24 DIMM
systems should have 12 DIMMs or 24 DIMMs installed). Fault Resilient Memory is disabled by default
and must be enabled through the BIOS setup menu.
Important: Consult your PowerEdge server installation and service manual for
complete memory population guidelines to properly enable Fault Resilient
Memory.
Memory Self-Healing
Memory Self-Healing Feature Support Table
DIMMs Supported
x4 DIMMs:
x8 DIMMs:
The Dell EMC PowerEdge server memory self-healing capability has two key components: post package
repair and memory retraining.
Post Package Repair (PPR) is an industry-standard capability, defined by JEDEC, where a memory module
is capable of swapping out degraded rows of memory with spare ones being held in reserve. While
JEDEC requires that all DDR4 memory be built with at least one spare row per DRAM bank group, Dell
requires all memory suppliers manufacture genuine Dell DIMMs with a significantly larger number of
available spare rows. This is done to ensure that PowerEdge servers have a robust self-healing memory
ecosystem.
FYI: Genuine Dell DIMMs are any memory modules procured directly from Dell
Technologies at time of system purchase, through aftermarket sales (e.g. dell.com),
or through Dell services as a replacement.
Any genuine Dell RDIMM or LRDIMM rated for speeds of 2666 MT/s or higher will
support the extended PPR resource requirements (extra spare rows) by Dell
Technologies.
When the server platform determines that a DRAM row has one or more faulty cells, it can instruct the
DRAM to electrically swap out the old row and replace it with a new one. This happens through
electrical fusing and is a permanent process. Additionally, the PPR process can only occur at the
beginning of a boot process – before memory training and test can occur. Similar to Memory Page
Retirement, deeming which DRAM require Post Package Repair is determined by a proprietary Dell
algorithm that takes into account correctable error rates and error patterns. As of BIOS versions 2.8.2 or
higher, self-healing through PPR on detection of uncorrectable errors is now also supported.