White Papers

ManualsBrandsDell ManualsConverged InfrastructureHigh Performance Computing Solution Resources

Comparing Haswell processor models for HPC applications

Garima Kochhar, September 2014

This blog evaluates four Haswell processor models (Intel® Xeon® E5-2600 v3 Product Family) comparing

them for performance and energy efficiency on HPC applications. This is part three in a three part series.

Blog one provided HPC results and performance comparisons across server generations, comparing Ivy

Bridge (E5-2600 v2), Sandy Bridge (E5-2600) and Westmere (X5600) to Haswell. The second blog

discussed the performance and energy efficiency implications of BIOS tuning options available on the

new Dell Haswell servers.

In this study we evaluate processor models with different core counts, CPU frequencies and Thermal

Design Power (TDP) ratings and analyze the differences in performance and power. Focusing on HPC

applications, we ran two benchmarks and four applications on our server. The server in question is part

of Dell’s PowerEdge 13

generation (13G) server line-up. These servers support DDR4 memory at up to

2133 MT/s and Intel’s latest E5-2600 v3 series processors (architecture code-named Haswell). Haswell is

a net new micro-architecture when compared to the previous generation Sandy Bridge/Ivy Bridge.

Haswell based processors use a 22nm process technology, so there’s no process-shrink this time around.

Note the “v3” in the Intel product name – that is what distinguishes a processor as one based on

Haswell micro-architecture. You’ll recall that “E5-2600 v2” processors are based on the Ivy Bridge micro-

architecture and plain E5-2600 series with no explicit version are Sandy Bridge based processors.

Haswell processors require a new server/new motherboard and DDR4 memory. The platform we used is

a standard dual-socket rack server with two Haswell-EP based processors. Each socket has four memory

channels and can support up to 3 DIMMs per channel (DPC).

Configuration

Table 1 below details the applications we used and Table 2 describes the test configuration on the new

13G server.

Table 1 - Applications and benchmarks

Application

Domain

Version

Benchmark

Stream

Memory bandwidth

v5.9

Triad

HPL

Computation - solve a dense

system of linear equations

From Intel MKL

Problem size 90% of total

memory

Ansys Fluent

Computational fluid

dynamics

v15.0

truck_poly_14m

LS-DYNA

Finite element analysis

v7_0_0_79069

car2car with endtime=0.02

WRF

Weather Research and

Forecasting

v3.5.1

Conus 2.5km

MILC

Quantum chromo dynamics

v7.7.3, v7.7.11

Input data file from Intel

Summary of content (6 pages)

PAGE 1
Comparing Haswell processor models for HPC applications Garima Kochhar, September 2014 This blog evaluates four Haswell processor models (Intel® Xeon® E5-2600 v3 Product Family) comparing them for performance and energy efficiency on HPC applications. This is part three in a three part series. Blog one provided HPC results and performance comparisons across server generations, comparing Ivy Bridge (E5-2600 v2), Sandy Bridge (E5-2600) and Westmere (X5600) to Haswell.
PAGE 2
Table 2 - Server configuration Components Server Details PowerEdge R730xd prototype Processor Memory Hard drive RAID controller 2 x Intel® Xeon® E5-2693 v3 – 2.6/2.2 GHz, 14c, 145W 2 x Intel® Xeon® E5-2680 v3 – 2.5/2.1 GHz, 12c, 120W 2 x Intel® Xeon® E5-2660 v3 – 2.6/2.2 GHz, 10c, 105W 2 x Intel® Xeon® E5-2640 v3 – 2.6/2.
PAGE 3
All the other applications show a steady performance improvement with higher bin processor models. For codes that have per-core licenses, the improvement with higher bin processors that have more cores is not commensurate with the increase in number of cores. For example, when comparing the 10c SKU to the 12c SKU, adding 20% more cores (20 cores vs. 24 cores) allows Fluent running truck_poly_14m a 17% performance improvement. Comparision across SKUs - Performance - ES.
PAGE 4
Comparision across SKUs - EE - ES.DAPC Relative Enerey Efficiency (higher is better) 1.20 1.11 1.00 0.97 0.80 1.00 0.96 1.07 1.00 0.98 1.01 1.03 1.001.02 1.01 1.04 1.00 1.04 0.95 0.991.00 1.05 1.04 1.03 0.991.00 0.95 0.991.00 0.94 0.88 0.60 0.40 0.20 0.00 Stream Triad HPL Fluent N=124320, NB=168 truck_poly_14m LS DYNA WRF MILC 7.6.3 MILC 7.7.11 car2car, endtime=0.02 Conus 2.5km Intel file Intel file E5-2697 v3 14c, 2.6/2.2 GHz, 145W E5-2680 v3 12c, 2.5/2.
PAGE 5
Comparision across SKUs - Performance, EE - COD.Perf Relative performance , EE (higher is better) 1.50 1.46 1.40 1.30 1.33 1.33 1.20 1.20 1.17 1.17 1.10 1.18 1.12 1.10 1.00 1.07 1.10 1.06 1.02 1.03 0.90 0.80 0.70 Stream HPL Fluent LS DYNA WRF MILC 7.6.3 MILC 7.7.11 Triad N=124320, NB=168 truck_poly_14m car2car, endtime=0.02 Conus 2.5km Intel file Intel file E5-2697 v3 14c, 2.6/2.2 GHz, 145W E5-2680 v3 12c, 2.5/2.1 GHz, 120W E5-2660 v3 10c, 2.6/2.
PAGE 6
Comparision across SKUs - idle and peak power - ES.DAPC Power consumption in Watts 600 500 1.33 400 1.10 1.00 300 0.88 200 100 1.03 1.03 1.00 1.00 0 Idle Peak E5-2697 v3 14c, 2.6/2.2 GHz, 145W E5-2680 v3 12c, 2.5/2.1 GHz, 120W E5-2660 v3 10c, 2.6/2.2 GHz, 105W E5-2640 v3 8c, 2.6/2.2 GHz, 90W Figure 4 - Power idle and peak - ES.DAPC Conclusion There is a clear performance up-side for all the applications and datasets studied here when using higher bin/higher core count processors.