Memory performance on HP Z840/Z640/Z440 Workstation

12345678
Number of cores accessing memory
Relative bandwidth
Figure 2. Memory bandwidth comparison
Channels populated vs. cores accessing memory
0.48
1.00
0.42
0.56
0.26
0.28
2 channels populated -
STREAM Triad
1 channels populated -
STREAM Triad
4 channels populated -
STREAM Triad
1.20
1.00
0.80
0.60
0.40
0.20
0.00
Figure 2 uses a STREAM subtest (Triad), to show how system memory bandwidth is impacted by the number of memory
channels loaded. For those not familiar with STREAM, it is a synthetic benchmark that measures the sustainable
memory bandwidth in a system. Figure 2 also shows how maximum memory bandwidth for a particular memory
channel loading is impacted by the number of cores accessing memory. Looking at the case of all the memory channels
loaded, it is apparent that the available memory bandwidth is not saturated until multiple cores are accessing memory
simultaneously.
Reviewing this relationship a general rule can be inferred. Single or lightly threaded applications that do not require or
use signicant memory bandwidth will experience a smaller performance impact moving to larger memory DIMMs, in
conjunction with populating fewer memory channels. Threaded applications that do require or use signicant memory
bandwidth will see a more signicant performance impact moving to larger memory DIMMs, in conjunction with
populating fewer memory channels.
Application/Benchmark performance comparisons
Application/benchmark performance comparisons were done on specic HP Z840/Z440 memory congurations, evaluating
the performance impact of moving to larger DIMM sizes in conjunction with populating fewer memory channels. The
comparisons look at moving from 4 GB DIMM congurations to 8 GB DIMM congurations, keeping the total memory
capacity equivalent. Similar performance results would be anticipated moving from 8 GB DIMM congurations to 16 GB
DIMM congurations, again keeping the total memory capacity equivalent.
The HP Z840 comparison results are representative of a dual processor HP Z640 conguration, while the HP Z440
comparison results are representative of a single processor conguration in either an HP Z640 or HP Z840 system.
System Processor Total
memory
Baseline
conguration
Comparison
conguration
HP Z440 Intel® Xeon®
E5-1680v3
16 GB 4 x 4 GB DDR4-2133 MHz
(4 channels loaded -
balanced conguration)
2 x 8 GB DDR4-2133 MHz
(2 channels loaded -
unbalanced conguration)
HP Z440 Intel® Xeon®
E5-1680v3
8 GB 2 x 4 GB DDR4-2133 MHz
(2 channels loaded -
unbalanced conguration)
1 x 8 GB DDR4-2133 MHz
(1 channel loaded -
unbalanced conguration)
HP Z840 Intel® Xeon®
E5-2687W
32 GB 8 x 4 GB DDR4-2133 MHz
(4 channels/proc loaded -
balanced conguration)
4 x 8 GB DDR4-2133 MHz
(2 channels/proc loaded -
unbalanced conguration)
HP Z840 Intel® Xeon®
E5-2687W
16 GB 4 x 4 GB DDR4-2133 MHz
(2 channels/proc loaded -
unbalanced conguration)
2 x 8 GB DDR4-2133 MHz
(1 channel/proc loaded -
unbalanced conguration)
Table 1. Memory congurations evaluated on the HP Z440 and Z840
Technical white paper | Memory performance on HP Z840/Z640/Z440 Workstations
3