Rank sparing system and method
A system, and a corresponding method, are used to implement rank sparing. The system includes a memory controller and one or more DIMM channels coupled to the memory controller, where each DIMM channel includes one or more DIMMS, and where each of the one or more DIMMs includes at least one rank of DRAM devices. The memory controller is loaded with programming to test the DIMMs to designate at least one specific rank of DRAM devices as a spare rank.
1. A system for implementing rank sparing, comprising:
a memory controller; and
one or more DIMM channels coupled to the memory controller, wherein each DIMM channel comprises one or more DIMMs, wherein each of the one or more DIMMs comprises at least one rank of DRAM devices, and wherein the memory controller comprises programming to test all the DIMMs (ranks) installed in the DIMM channels to designate at least one specific rank of DRAM devices as a spare rank.
2. The system of claim 1 , wherein the memory controller comprises:
threshold programming to determine if an error threshold is reached in any active DIMM ranks; and
rank test programming to test viability of all DIMM ranks upon boot up of a corresponding computer system.
3. The system of claim 1 , wherein the DIMMs installed in the DIMM channels each are 1-rank DIMMs.
4. The system of claim 1 , wherein the DIMMs installed in the DIMM channels each are 2-rank DIMMs.
5. The system of claim 1 , wherein the DIMMs installed in the DIMM channels comprise 1-rank DIMMs and 2-rank DIMMs.
6. The system of claim 1 , where in the DIMMs installed in the DIMM channels comprise 2-rank DIMMs and 4-rank DIMMs.
7. The system of claim 1 , where in the DIMMs installed in the DIMM channels are 4-rank DIMMs.
8. The system of claim 1 , wherein the DIMMs installs in the DIMM channels comprise 1-rank, 2-rank and 4-rank DIMMs.
9. The system of claim 1 , wherein the memory controller comprises programming to issue an alert to a user when a rank is not available for sparing.
10. The system of claim 1 , wherein rank sparing is enabled as an optional feature.
11. A method, implemented in a memory-subsystem of a computer, for rank sparing, the method, comprising:
at a memory controller of the memory sub-system, receiving a boot-up notification;
in response to the boot-up notification, testing each rank of memory in the memory sub-system to determine availability of ranks for active operations and sparing operations; and
designating one or more specific ranks as spare ranks, wherein operation with rank sparing increases reliability of the memory sub-system.
12. The method of claim 11 , wherein the memory sub-system, comprises:
a memory controller;
one or more DIMM channels coupled to the memory controller, wherein the DIMM channels comprise one or more channel DIMMs, and wherein each of the one or more channel DIMMs comprises at least one rank of DRAM devices, where in the memory controller comprises programming to test the channel DIMMs to designate at least one specific rank of DRAM modules as a spare rank.
13. The memory of claim 11 , wherein the memory controller determines an insufficient number of ranks for both sparing and active operations, the method further comprising:
sending a warning message; and
implementing memory operation without rank sparing.
14. The method of claim 11 , further comprising:
monitoring operation of the memory sub-system for occurrence of correctable faults;
comparing a determined number of correctable faults to a threshold value; and
when the threshold value is reached, enabling one or more spare ranks.
15. The method of claim 14 , wherein enabling one or more spare ranks, comprises:
copying the memory contents from a failing rank to one of the spare ranks; and
redirecting memory transactions from the failing rank to the one of the spare ranks.
16. The method of claim 12 , wherein the channel DIMMs each are 1-rank DIMMs.
17. The method of claim 12 , wherein the channel DIMMs each are 2-rank DIMMs.
18. The method of claim 12 , wherein the channel DIMMs comprise 1-rank DIMMs and 2-rank DIMMs.
19. The method of claim 12 , wherein the channel DIMMs comprise 2-rank DIMMs and 4-rank DIMMs.
20. The method of claim 12 , wherein the channel DIMMs are 4-rank DIMMs.
21. The method of claim 12 , wherein the channel DIMMs comprise 1-rank, 2-rank and 4-rank DIMMs.
22. The method of claim 12 , further comprising:
determining if sufficient additional ranks are available for sparing and active memory operations; and
is sufficient ranks are available, re-initiating rank sparing.
23. A computer memory enabling rank sparing, comprising:
a plurality of memory devise arranged into two or more ranks; and
a memory controller coupled to each of the two or more ranks, the controller, comprising:
means for inventorying and testing the memory devices to determine an availability of one or more ranks for rank sparing,
means for designating one or more ranks for sparing, and
means for implementing rank sparing using the one or more designated ranks.