![]() |
|
![]() |
|
Nehalem Memory Speed InformationThe memory controller in Intel's new Nehalem processor is a radical departure from previous Intel processor architectures, in that the processor die now contains a memory controller. As a result of this change, much of the system builder community is confused as to the proper memory speed to use for Nehalem processors. This paper provides information to system builders so that the proper memory can be selected.
This
information is extracted from Intel documentation. It is correct,
to the best of our knowledge, as of the time it was written.
|
|
When putting together a Nehalem system, the following configuration parameters should be followed to ensure maximum memory bandwidth.
For Dual-Processor systems:
For ALL systems:
It should be noted that memory bandwidth had no significant impact on the benchmark we used above (Cinebench), but there are workloads where memory bandwidth will have an effect. The most common is virtualization. Additionally, some HPC (High Performance Computing) workloads are VERY sensative to memory bandwidth.
In addition to the links above, the Intel Xeon Processor 5500 Series Datasheet, Volume 2, specifically chapter 3, was very useful here, along with Intel's Specification Updates for the various processors.
Some information from the Dell Enterprise Center - TechCenter Wiki - Nehalem and Memory Configurations was useful.
The following table lists the motherboards that were used in this project.
Manufacturer |
Model |
Type |
Intel |
S5520HC |
Dual-Socket |
Intel |
S5520UR |
Dual-Socket |
Supermicro |
X8STE |
Single-Socket |
The following table lists the memory that was used in this project. There were no issues with any one type of this memory in the boards. As there were occasional minor stability issues when memory was mixed, it is suggested that only a single model number of memory be used in a system.
The Registered memory was not used in the single-socket system.
Manufacturer |
Model |
Type |
Clock |
Rank |
Capacity |
CAS at 666MHz |
CAS at 533MHz |
Crucial |
CT12872BA1339.9SDF |
Unbuffered ECC |
666 MHz |
Single |
1GB |
9 |
7 |
Kingston |
KVR1066D3E7/2G |
Unbuffered ECC |
533MHz |
Dual |
2GB |
N/A |
7 |
Kingston |
KVR1333D3E9/2G |
Unbuffered ECC |
666MHz |
Dual |
2GB |
9 |
8 |
Kingston |
KVR1333D3D4R9S/4G |
Registered ECC |
666MHz |
Dual |
4GB |
9 |
??? |
Super-Talent (see note) |
W1333EA1GM |
Unbuffered ECC |
666MHz |
Single |
1GB |
9 |
8 |
Super-Talent (see note) |
W1333EB2GM |
Unbuffered ECC |
666MHz |
Dual |
2GB |
9 |
8 |
Super-Talent (see note) |
W13RB4G4H |
Registered ECC |
666MHz |
Dual |
4GB |
9 |
8 |
Super-Talent (see note) |
W13RB4G4Q |
Registered ECC |
666MHz |
Dual |
4GB |
9 |
7 |
NOTE: Although the Super-Talent memory worked correctly where used in these tests, other testing has shown incompatabilities. For this reason, combined with the overall poor support we received from the manufacturer, we do not suggest or advise that Super-Talent memory be used for new systems.
| Prev | Index |
If you have any comments or suggestions, please E-mail webmaster@accs.com