<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
</head>
<body bgcolor="#FFFFFF" text="#000000">
You'd think that by now I'd know better. Trying to live on the
cutting edge. But the promise of 5% over Haswell was quite
alluring. <br>
<br>
We purchased Broadwell 120W 2680v4 chips with 128G of RAM enclosed
in the Dell M630 blades. When we finally received power the first
thing we did was load a RHEL7 OS, checked BIOS to be sure we had all
the performance variables set and ran HPL compiled with Intel v16
compilers against their MKL. Performance went from a high of 791
GFLOPS down to 679. A whopping 14% difference.<br>
<br>
Test element low high % mean
median std dev <br>
WR00L2L4 315 679.44 791.34 14.14
741.67 739.64 19.89 <br>
<br>
We should be able to do better than this and reduce to something
like 5% right?<br>
<br>
We checked power settings for the chassis and played with those. we
turned power management to BIOS and then to the OS using ACPI. No
difference. We swapped the fastest and slowest nodes thinking that
this might be a location issue. No difference. And then we found a
BIOS update from 2.0.1 to 2.1.6 which was fresh so loaded that one
up.<br>
<br>
Performance went down. Considerably.<br>
<br>
Test chassis low high % mean
median std dev <br>
WR00L2L4 296 583.47 636.27 8.30 623.74
625.15 9.82 <br>
<br>
Wow! 741 to 623 GFLOPS!<br>
<br>
We then looked at power and heat using the turbostat command to log
values. What we found was that at the slowest nodes the SMI
interrupts and c1 states were higher and the power was capped at
120W. On the fastest nodes, things were different with power
hovering around 117W. Again switching node slots changed nothing.<br>
<br>
With Dell's help we finally manged to turn off turbo mode and set
the --ProcConfigTdp=Level1 to only run at the base AVX speed of
1.9GHz. This indeed provided much closer HPL results.<span
style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D"></span>
<br>
<br>
Test chassis low high % mean
median std dev <br>
WR00L2L4 310 515.95 519.55 0.69 519.07
519.14 0.34 <br>
<br>
with plenty of the nodes hovering around the 110W usage.<br>
<br>
But now we had a new motherboard and with the same setup ran another
test, this time without updating BIOS so it still was sitting on
2.0.1. Lo and behold, there's that (weak) performance again. 638
GFLOPS and better power usage.<br>
<br>
638.12, Temps: 78, 70, Watts: 107.02, 104.65<br>
<br>
We still don't understand what we are up against here. Obviously
re-enabling the performance variables in BIOS will begin to get
those FLOPS up again. As will degrading the BIOS and microcode.
And maybe this 14% difference between best and worst nodes is all we
can expect. But I'd sure like to have a lot more of those better
performers!<br>
<br>
Did I mention GPFS? We have it running on a v3 node with the same
kernel. On the Broadwell chips though, it just hangs the kernel.
Sigh. The cutting edge. When can I order Skylake?<br>
<br>
Bill<br>
</body>
</html>