Сегодня покупают
← Вернуться назад
Нужна помощь с конфигурацией?
Пришлите SKU, спецификацию или список требований. Специалист Apltech проверит совместимость компонентов и найдёт актуальные артикулы
Получить консультацию →

HPE ProLiant Benchmark Newsletter: технические характеристики и документация

HPE Server Performance Benchmarks

 

HPE Server Performance Benchmarks

Newsletter

 

September 2019

 

HPE Gen10 servers SHATTER with 37 world records on AMD EPYC™ leaderboard!1

 

UPDATE: 191 world records

http://www.crn.com/sites/default/files/ckfinderimages/userfiles/images/crn/misc/2017/hpe-discover-gen10-400.jpg

HPE competitive benchmarking results give an important edge to successful sales techniques. HPE servers sustain world record status in performance and price-performance. The HPE portfolio of ProLiant Gen10 servers, HPE Integrity servers, and HPE Synergy Gen10 Compute Modules carry on, achieving worldwide #1 records across the performance landscape. See Marketing Documents Library webpage for Benchmark and Performance Guides.

The HPE ProLiant DL385 Gen10 and DL325 Gen10 recently took a total of 72 performance leadership results and 16 First records on the SPEC CPU 2017 benchmark. HPE Gen10 servers now hold 191 world records and 47 First records!

11 more records than Lenovo and more than Alibaba, Giga-Byte, and Supermicro COMBINED!

HPE ProLiant DL325 Gen10

 

HPE Performance Leadership!

 

192

WORLD RECORDS

 

 

HPE Performance Trailblazer!

 

47

FIRSTS

 

SPECjbb2015
Composite critical-jOPS

1-processor (1P) world record

1P world record on Linux®

SPECjbb2015 
Composite max-jOPS

1P world record

1P world record on Linux

SPECjbb2015
Distributed critical-jOPS

1P world record

1P world record on Linux

SPECjbb2015
MultiJVM max-jOPS

1P world record

1P world record on Linux

Energy efficiency:
SPECpower_ssj2008

1S world record on Windows

TPC-H

1P performance world record @ 1TB (non-clustered)

1P overall price-performance world record @ 1TB (non-clustered)

1P, 4-node is 4P world record (historical) cluster @ 1TB

1P, 8-node overall price-performance world record (tie) @ 1TB

1P, 8-node overall price-performance world record (tie)

1P, 8-node is 8P cluster world record (historical) cluster

TPC Express Benchmark ™ -HS (TPCx-HS)

1P Overall 10TB world record @ 10TB

1P 10TB world record @ 10TB

1P overall price-performance @ 10TB world record

1P overall 30TB world record @ 30TB

1P, 1S 30TB world record @ 30TB

1P, 1S 30TB overall price-performance world record @ 30TB

VMmark® 3.1

1S, 4-host world vSAN record on AMD EPYC 7702

HPE ProLiant DL385 Gen10

SPECjbb2015
Distributed critical-jOPS

2P world record

2P world record on Linux

SPECjbb2015
Distributed max-jOPS

2P world record

2P world record on Linux

SPECjbb2015 MultiJVM max-jOPS

2P world record on Linux

SPECpower_ssj2008

2S world record

2S world record on Linux

2S world record on Windows

Virtualization:
SPECvirt_sc2013

2P world record

TPCx-V

Overall world record

2P world record

Overall price-performance world record

VMmark 3.1

2-host, matched pair 2S world record

SAP Sales and Distribution standard application benchmark, two-tier

2P world record

2P world record on Windows

AMD EPYC Leaderboard records and competitive claims for listed benchmark results are as-of August 7, 2019, unless otherwise noted.


1https://www.amd.com/en/processors/epyc-world-records

 

See next page for more wins and performance details.

HPE Gen 10 servers continue to hold numerous leadership results!

 

The following lists current world records and industry-first successes with the latest and prior generation Intel® Xeon® and AMD EPYC processors. Find detailed slides on these records in the Performance wins deck that is viewable internally by HPE and Partners. For a complete list of performance briefs across all processor generations, see the Marketing Documents Library

 

ProLiant DL325

ProLiant DL385

HPE ProLiant DL360 Gen10 Server Center facing horizontal

HPE ProLiant DL385 Gen10 7251 2.1GHz 8-core 1P 16GB-R P408i-a 8SFF 500W PS Server

Performance Score Card

SPECpower_ssj2008

  • The most energy-efficient 1U server on Windows!
  • The most energy-efficient 1P server on Windows!
  • First server efficiency score on Microsoft Windows 2019!
  • First 1P and 1U AMD-based server to
    break 8,000 overall ssj_ops/watt

TPCx-HS

On the TPCx-HS @ 10TB scale factor:

  • #1  overall and #1 1P performance
  • #1 overall price-performance

On the TPCx-HS @ 30TB scale factor:

  • #1 overall and #1 1P performance
  • #1 overall price-performance

TPC-H @ 1000GB

Non-clustered:

  • #1 and first 1P performance
  • #1 overall price-performance

4-node cluster:

  • #1 1P performance for a 4-node cluster

8-node cluster:

  • #1 overall performance for clustered systems
  • #1 1P performance for an 8-node cluster

4-node and 8-node clusters (tied):

  • #1 overall price-performance for clustered systems

VMmark 3.1

  • #1 1P 4-node result
  • First 1P 4-node result

SPECjbb2015

  • #1 1P MultiJVM max-jOPS
  • #1 1P Distributed max-jOPS
  • #1 1P Composite max-JOPS
  • #1 1P Composite critical-jOPS

SPEC CPU 2017*

  • Most energy-efficient 1U server in all 8 categories
  • Most energy-efficient 1P server in all 8 categories
  • Most energy-efficient 1U server on Linux in all 8 categories
  • Most energy-efficient 1P server on Linux in
    all 8 categories
  • Overall most energy-efficient server in all 4 SPECspeed2017 categories
  • First 1P results on all 8 SPEC CPU 2017 energy metrics:

For SPECrate2017

For SPECspeed2017

-  int_energy_base > 800

-  int_energy_peak > 875

-  fp_energy_base > 700

-  fp_energy_peak > 775

- int_energy_base > 43.0

- int_energy_peak > 45.0

- fp_energy_base > 375

- fp_energy_peak > 375

Performance Score Card

SPECpower_ssj2008

  • Most energy-efficient 2P server on the planet
  • Most energy-efficient 2P server on Linux
  • Most energy-efficient 2P server on Windows
  • Most energy-efficient 2U server on Windows
  • First 2P result that beat the 18,000 server-efficiency barrier
  • First server-efficiency score on Microsoft Windows 2019

SPECjbb2015

  • #1 2S Distributed max-jOPS
  • #1 2S Distributed critical-jOPS

SPECvirt_sc2013

  • #1 2P virtualization performance
  • #1 2P price-performance

TPCx-V

  • #1 overall performance and price-performance

VMmark 3.1

  • #1 2P 2-node performance result

SAP Sales and Distribution standard application benchmark, two-tier

  • #1 overall 2P performance

SPEC CPU 2017*

  • Most energy-efficient server in all 8 categories of SPEC CPU 2017 metrics for:
  • 2P server and 2U server
  • 2P server and 2U server on Linux
  • Most energy-efficient 2P server on Linux
  • Overall most energy-efficient server in all 4 SPECrate2017 categories
  • First 2P results in all 8 SPEC CPU 2017 energy metrics:

For SPECrate2017

For SPECspeed2017

-  int_energy_base > 1080

-  int_energy_peak > 1190

-  fp_energy_base > 861

-  fp_energy_peak > 927

-  int_energy_base > 40.9

-  int_energy_peak > 43.8

-  fp_energy_base > 347

-  fp_energy_peak > 348

 

*SPEC CPU 2017 as-of-date is September 18, 2019

 

 


 

AWARDS and RECOGNITION

 

ProLiant DL325 Gen10
IT Pro Editor's Choice

An EPYC-powered stunner

Product of the Year

HPE ProLiant DL325 Gen10

Favorite pick for All-Purpose Server

ProLiant DL385 Gen10
IT Pro Editor's Choice

Big power, bigger value

 

ProLiant DL380

ProLiant DL560

HPE ProLiant DL560 Gen10 Server Center facing

Performance Score Card

VMmark

  • #1 2-node 2P VMmark 2.5.2
  • First published benchmark on VMmark 3.0 and 2.5.x with Intel Xeon Scalable Processors

TPC-H (non-clustered)

  • #1 2P @ 3000GB performance and price-performance
  • #1 2P Linux  @ 3000GB performance and price-performance
  • #1 overall and 2P Microsoft SQL Server @ 3000GB performance and price-performance
  • #1 overall and 2P Microsoft Windows @ 3000GB performance and price-performance
  • #1 2P performance and price-performance @ 1000GB

Performance Score Card

SAP Business Warehouse, Edition for SAP HANA Standard Application Benchmark Version 3

  • 6.5 billion initial records:

- #1 4P on Phase 1

  • 5.2 billion initial records:

- #1 4P on Phase 1

- #1 4P on Phase 3

SPECjbb2015

  • #1 4P max-jOPS
  • #1 4P/2U Linux MultiJVM max-jOPS (tie)
  • #1 4P/2U MultiJVM max-jOPS
  • #1 4P/2U Distributed max-jOPS and critical-jOPS

SPECpower_ssj2008

  • First 4P server with second-generation Intel Xeon processors to beat 13,000 overall ssj_ops/watt

ProLiant ML350

ProLiant DL580

Performance Score Card

SPECpower_ssj2008

  • #1 overall energy-efficient tower
  • #1 2P energy-efficient tower
  • World's MOST energy-efficient tower
    Linux-based
  • World's MOST energy-efficient tower
    Windows-based
  • #1 2P tower Linux-based
  • #1 2P tower Windows-based
  • #1 tower using Intel Xeon Scalable Processor family

Performance Score Card

SPECpower_ssj2008

  • #1 4P/4U Linux
  • First 4P/4U to beat 12,000 overall ssj_ops/watt

 

AWARDS and RECOGNITION

 

IT Pro Editor's Choice

HPE ProLiant DL560 Gen10
IT Pro Editor's Choice

High density Xeon Scalable power station at a tempting price

IT Pro Editor's Choice

HPE ProLiant ML350 Gen10
IT Pro Editor's Choice

A perfect balance of power and  value

 

Synergy 480

Synergy 660

HPE Synergy 480 Gen10 Compute Module Right facing

Performance Score Card

SPECpower_ssj2008

  • #1 12-node
  • #1 12-node on Microsoft Windows
  • #1 12-node on Linux
  • #1 2P Windows: 5-node, 6-node, 7-node, 8-node, 9-node, 10-node, 11-node, 12-node
  • First 12-node system to beat the 11,000 barrier
  • First 2P system published with the Meltdown and Spectre security patches, variants #1 and #2
  • First 2P Linux Composable Infrastructure

VMmark

  • First and #1 Composable Infrastructure on VMmark 3.0
  • for the 6-node result

- #1, first, and only 6-node/2P result

- #1 6-node composable infrastructure result

  • for the 8-node result:

- #1 and first 8-node/2P result

- #1 8-node composable infrastructure result

  • for the 16-node result:

- #1 and first 16-node result

- #1 and first 16-node/2P result

- #1 and first 16-node composable infrastructure result

  • for the 24-node result

- #1 overall result

- #1 and first 24-node result

- #1 and first 24-node/2P result

- #1 and first 24-node composable infrastructure result

Performance Score Card

SPECpower_ssj2008

  • #1 4P Windows multi-node
  • #1 2-node, 3-node, 5-node, 6-node
  • #1 4P 2-node, 3-node, 4-node, 5-node, 6-node
  • #1 4P Windows: 2-node, 3-node, 4-node, 5-node, 6-node
  • #1 4P Linux: 2-node, 3-node, 4-node, 5-node, 6-node
  • First 4P Linux Composable Infrastructure
  • First multi-node to beat 12,000 overall ssj_ops/watt
  • First 4P multi-node with over 11,000 ssj_ops_watt
  • First 4P system published with the Meltdown and Spectre security patches, variants #1 and #2

 

 

Superdome Flex

Performance Score Card

SPEC CPU2017 benchmark

  • #1 overall, 32P and 16P SPECrate2017_int_base and SPECrate2017_fp_base

SPEC CPU2006 benchmark

  • #1 32P SPECint_rate_base2006
  • #1 32P SPECfp_rate_base2006
  • #1 16P SPECint_rate_base2006
  • #1 16P SPECfp_rate_base2006
  • First and only published SPEC CPU2006 results with 32 and 16 Intel Xeon Scalable processors

SPEC OMP2012 benchmark

  • Four world records for  overall, 32P, 16P, and 8P SPECompG_base2012
  • Four world records for overall, 32P, 16P, and 8P SPECompG_peak2012

SAP Business Warehouse, Edition for SAP HANA Standard Application Benchmark Version 3

  • 11.7 billion initial records:

-  #1 8P on Phase 1

-  #1 8P on Phase 3

  

Fair use and compliance information and system configurations

All benchmark data is publicly available on the websites listed below. All performance briefs listed are publicly available on hpe.com or on the server Documents webpages and are customer viewable. See benchmark details below.

HPE Configurations

SAP® Business Warehouse (BW), edition for SAP HANA® standard application benchmark version 3

ProLiant DL385 Gen10:

Claim based on overall #1 2P performance on the two-tier SAP Sales and Distribution standard application benchmark with AMD EPYC 7702 2.20 GHz processors; 2 processors/128 cores/256 threads; Microsoft Windows Server 2019; Microsoft SQL Server 2017; SAP enhancement package 5 for SAP ERP 6.0; 1 TB main memory. Certification #2019044. Number of benchmark users: 45,600; SAPS: 249,550; Average dialog response time: 0.96 seconds. Tested by HPE on August 7, 2019.

ProLiant DL560 Gen10:

Claim based on best-published 4P SAP HANA V3 Standard Application Benchmark @ 5.2 billion (5.2B) initial records result published as of April 2, 2019. 4 Intel Xeon Platinum 8280L processors (112 cores/224 threads) with 3072-DDR4 GB total memory, running SUSE Linux Enterprise Server (SLES) 12 SP4, using HANA 2.0 Rev. 35 and NetWeaver® 7.50. Certification# 2019019. SAP HANA V3 Standard Application Benchmark score: Runtime of Data Load/Trans (14,030 secs), Query Executions per Hour (5086 secs), Total Runtime of Complex Query (97 secs) @ 5.2B. Tested by HPE on March 22, 2019.

Claim based on best-published 4P SAP HANA V3 Standard Application Benchmark @ 6.5 billion (6.5B) initial records result published as of April 2, 2019. 4 Intel Xeon Platinum 8280L processors (112 cores/224 threads) with 3072-DDR4 GB total memory, running SLES12 SP4, using HANA 2.0 Rev. 35 and NetWeaver 7.50. Certification# 2019022. SAP HANA V3 Standard Application Benchmark score: Runtime of Data Load/Trans (18821 secs), Query Executions per Hour (secs), Total Runtime of Complex Query (107 secs) @ 6.5B. Tested by HPE on March 22, 2019.

Claim based on best published 4P at 6.5 billion initial records with HPE Persistent Memory result published as of April 2, 2019. 4 Intel Xeon Platinum 8280L 2.70 GHz processors with (112 cores/224 threads) with 32 GB HPE DDR4 Smart Memory and HPE Persistent Memory total running SLES 12 SP4 using SAP HANA 2.0 Rev. 35 and NetWeaver 7.50 SP04. Certification #2019020. SAP HANA V3 Standard Application Benchmark score: Runtime of Data Load/Trans (21,533 secs), Query Executions per Hour (3,825 secs), Total Runtime of Complex Query (135 secs) @ 6.5B.

Superdome Flex

Claim based on 11.7 billion initial records: 8 Intel Xeon Platinum 8280L 2.70 GHz processors with (224 cores / 448 threads) with 12 TB memory; SUSE Linux Enterprise Server 12 SP 4, SAP HANA 2.0, NetWeaver 7.50. Source: Certification number 2019021: Scores: Total Runtime of Data Load for the last data set/Transformation 20,947 seconds; Query Executions per Hour 4,722; Runtime of Complex Query (150 secs) @ 11.7B. Results achieved with Intel security mitigations are not comparable to results achieved without the mitigations.

TPC-H

ProLiant DL325 Gen10:

Non-clustered result: Claim based on having the #1 price-perform and #1 and first 1P performance for a non-clustered system on the TPC-H @ 1000GB scale factor. Configuration: 1 HPE ProLiant DL325 Gen10 server used 1 AMD EPYC 7502P 2.5 GHz processor; 1 socket/32 cores/64 threads; Red Hat® Enterprise Linux 8; Microsoft SQL Server 2017 Enterprise Edition. TPC-H results show the HPE ProLiant DL325 Gen10 with a result of 743,750.60 QphH @ 1000GB and $0.34 USD/QphH @ 1000GB with a system availability of August 26, 2019; see tpc.org/3342.

4-node and 8-node results: Claim based on having the #1 1P performance for a 4-node cluster; #1 overall performance for clustered systems; #1 1P performance for an 8-node system; and #1 overall price-performance for clustered systems (tie for 4-node and 8-node clusters) on the TPC-H @ 1000GB scale factor.
Configuration:
4 nodes: 4 HPE ProLiant DL325 Gen10 servers used 1 AMD EPYC 7702 2.0 GHz processor each; total 4 socket/256 cores/512 threads; CentOS 7.6, Exasol 6.2. TPC-H results show the HPE ProLiant DL325 Gen10 with a result of 3,635,443 QphH @ 1000GB and $0.06 USD/QphH @ 1000GB with a system availability of August 26, 2019; see tpc.org/3340; 8 nodes: 4 HPE ProLiant DL325 Gen10 servers used 1 AMD EPYC 7702 2.0 GHz processor each; total 4 socket/512 cores/1024 threads; CentOS 7.6, Exasol 6.2. TPC-H results show the HPE ProLiant DL325 Gen10 with a result of 6,145,628 QphH @ 1000GB and $0.06 USD/QphH @ 1000GB with a system availability of August 26, 2019; see tpc.org/3341. Benchmark results as of August 8, 2019.

ProLiant DL380 Gen10:

Claim based on best-published 2P TPC-H benchmark @ 3,000GB non-cluster results published at tpc.org as of April 2, 2019. New configuration: 1-Node, 2 Intel Xeon 8280 processors (2 processors, 56 cores, 112 threads) with 1,536 GB total memory on SUSE Linux Enterprise Server 15 using Microsoft SQL Server 2017 Enterprise Edition. Source: tpc.org/3336. Score: 1,244,450.6 QphH @ 3000GB, $0.38/QphH @ 3000GB available April 2, 2019. Tested by HPE on April 1, 2019.

Claim based on best published TPC Express Benchmark Big Bench (TPCx-BB) @ 10000 GB result published at www.tpc.org as of April 2, 2019. New configuration: 18 HPE ProLiant DL380 Gen10 servers with 2 Intel Xeon Gold 6254 processors and 3 ProLiant DL360 Gen10 servers with 2 Intel Xeon Gold 5218 processor for worker nodes on Red Hat Enterprise Linux Server 7.5 on Cloudera for Apache Hadoop (CDH) 5.16.1. Source: tpc.org/3510. Score: 1,759.64 BBQpm @10000 @ $521.34 USD/BBQpm @ 10000 GB available April 2, 2019. Tested by HPE on 1 April, 2019

1000GB scale factor

Claim based on the #1 world record with a non-clustered result of 1,009,065 QphH @ 1000GB and $0.47 USD/QphH @ 1000GB with system availability of May 21, 2018; see tpc.org/3331.

3000GB scale factor

Claim based on the #1 record with a non-clustered result of 1,244,451 QphH @ 3000GB and $0.38 USD/QphH @ 3000GB with system availability of April 2, 2019; see tpc.org/3333.


 

TPC-HS

ProLiant DL325 Gen10:

Claim based on having the #1 overall performance and price-performance @ 10TB scale factor and the #1 overall performance and price-performance @ 30TB scale factor. Configurations: 17 HPE ProLiant DL325 Gen10 servers with AMD EPYC 7502P 2.0 GHz processors; 17 sockets/544 cores/1088 threads; Red Hat Linux Enterprise Server 7.6; Hortonworks Data Platform 3.1; 256 GB memory. Results: 23.66 HSph@10TB and $25,057.91 $/HSph@10TB; system availability August 26, 2019; see tpc.org5532; 17 HPE ProLiant DL325 Gen10 with AMD EPYC 7502P 2.0 GHz processors; 17 sockets/544 cores/1088 threads; Red Hat Linux Enterprise Server 7.6; Hortonworks Data Platform 3.1; 256 GB memory. Results: 25.67 HSph@30TB and $27,649.40 $/HSph@30TB; system availability August 26, 2019; see tpc.org/5533.

TPCx-V

ProLiant DL385 Gen10:

Claim based on #1 overall performance and price-performance TPCx-V benchmark published at tpc.org as of August 7, 2019. Configured as AMD EPYC 7702 2.0 GHz processors: 2 processors/ 128 cores/ 256 threads; Red Hat Enterprise Linux Server 7.6; VMware® vSphere® 6.7; system availability October 15, 2019. Scores: 2,280 tpsV and $36.26 USD per tpsV. See tpc.org/5302

VMmark® 3.1

ProLiant DL325 Gen10:

Claim based on having the #1 1P 4-node result and the first P 4-node result with AMD EPYC 7702 processors with a score of 12.23 @ 13 tiles. Results as of August 7, 2019.

ProLiant DL385 Gen10:

Claim based on being the #1 2P 2-node result with AMD EPYC 7702 processors with a score of 2.78 @ 14 tiles. Results as of August 7, 2019.

VMmark 3.0

Synergy 480 Gen10:

Claim based on being the first and #1 composable infrastructure result with Intel Xeon Platinum 8180 processors and with a score of 7.93@ 8 tiles.

Claim based on being the first published results and the first results with Intel Xeon Platinum 8180 processors with a score of 5.88 @ 6 tiles.

Claim based on having the #1, first, and only 6-node/2P results; and #1 6-node composable infrastructure result with Intel Xeon Gold 6252 processors and with a score of
16.91 @ 18 tiles.

Claim based on having the #1 and first 8-node/2P result and #1 8-node composable infrastructure result with Intel Xeon Gold 6140 processors and with a score of
15.25 @ 16 tiles.

Claim based on having the #1 and first 16-node result; the #1 and first 16-node/2P result, and #1 and first 16-node composable infrastructure result with Intel Xeon Gold 6140 processors and with a score of 30.13 @ 32 tiles.

Claim based on having the #1 overall result; the 1 and first 24-node result; #1 and first 24-node/2P result, and #1 and first 24-node composable infrastructure result with Intel Xeon Gold 6140 processors and with a score of 46.10 @ 48 tiles.

ProLiant DL580 Gen10:

Claim based on having the first 2-node, 4P result on the VMmark 3.0 benchmark; score 14.82 @ 15 tiles.

VMmark 2.5.2

ProLiant DL380 Gen10:

Claim based on the #1 2P, two-node score on the VMmark 2.5.2 benchmark with a score of 39.75 @ 34 tiles which is also the first result with Intel Xeon Platinum 8180 processors on the VMmark 2.5.2 benchmark.

SPEC OMP2012

For SPECompG_peak2012, various thread counts were used for the different socket configurations with a range of 144 to 288 threads for the 8S result, a range of 256 to 576 threads for the 16S result, and a range of 512 to 576 threads for the 32S results.

 

 

 

Sign up for updates

 

 

© Copyright 2019 Hewlett Packard Enterprise Development LP. The information contained herein is subject to change without notice. The only warranties for Hewlett Packard Enterprise products and services are set forth in the express warranty statements accompanying such products and services. Nothing herein should be construed as constituting an additional warranty. Hewlett Packard Enterprise shall not be liable for technical or editorial errors or omissions contained herein.

 

AMD is a trademark of Advanced Micro Devices, Inc. Intel Xeon is a trademark of Intel Corporation in the U.S. and other countries. Linux is the registered trademark of Linus Torvalds in the U.S. and other countries. Microsoft, Windows, and Windows Server are either registered trademarks or trademarks of Microsoft Corporation in the United States and/or other countries. Red Hat is a registered trademark of Red Hat, Inc. in the United States and other countries. SAP and SAP HANA are trademarks or registered trademarks of SAP SE (or an SAP affiliate company) in Germany and other countries. VMmark, VMware vSAN, VMware vSphere, and VMware are registered trademarks or trademarks of VMware, Inc. and its subsidiaries in the United States and other jurisdictions. SPEC and the benchmark names SPEC CPU, SPECfp, SPECint, SPECrate, SPECspeed, SPECjbb, SPECpower_ssj, SPEC OMP, and SPEC VIRT_SC are registered trademarks of the Standard Performance Evaluation Corporation (SPEC). All rights reserved; see spec.org. TPC, TPC-H, TPC-HS, and TPCx-V are trademarks of the Transaction Processing Performance Council. The TPC believes that comparisons of results published with different scale factors and currencies are misleading and discourages such comparisons. All third-party marks are property of their respective owners.

 

a50000274enw, 14777  September 2019