Dell Fault Resilient Memory - Virtualization - Wiki - Virtualization - Dell Community

Dell Fault Resilient Memory

Virtualization - Wiki

Wikis - Page

Dell Fault Resilient Memory

Virtualization - Wiki

Introduction to Dell Fault Resilient Memory (FRM)

Fault Resilient Memory (FRM) is a new Memory Operating Mode available on Dell PowerEdge 12th generation servers. The mode establishes an area of memory that is fault resilient and can be used by an Operating System supporting the FRM  feature to load critical Applications/Services  & the OS Kernel to maximize system availability. VMware ESXi 5.5 supports FRM through its Reliable Memory (ReM) Feature. 

Benefits of enabling FRM

            FRM creates a highly resilient memory zone for the hypervisor, protecting it from all but the most severe of memory errors.  With the VMware Reliable Memory feature, vSphere v5.5 is able to take advantage of this zone, providing the hypervisor with a strong protection from memory faults that would otherwise bring down the entire system.  When used in conjunction with other Dell PowerEdge reliability and redundancy features for virtualization -- such as the Internal Dual SD Modules usable with ESXi installations and Memory Page Retire, which helps the hypervisor to locate and fence off areas prone to many correctable errors that could lead to an uncorrectable error in the standard memory space -- it creates a highly reliable VMware virtualization server environment for customers concerned with high availability and uptime.

For more details, refer to white paper posted at Dell tech center. 

Comments
  • VERY difficult to find actual numbers on how much more efficient FRM is over MM (memory mirroring).  Turns out FRM uses 12.5% available memory, while MM uses 50%.

  • In the case of FRM, redundancy is at the memory channels level. With that, I believe it's easy to calculate the space used for FRM.