Home > Blogs > VMware VROOM! Blog > Monthly Archives: April 2010

Monthly Archives: April 2010

Three-Tier SAP SD Benchmark on vSphere 4 Supports 16,000 Users

The first SAP Three-Tier SD benchmark on vSphere 4.0 was recently completed and published by Fujitsu.  The published results of this test highlight support of an impressive 16,000 users with a SAPS rating of 87,800.  This shows that large numbers of SAP users can be supported in a virtual environment spread across many hosts while providing good performance. 

Eight Fujitsu Primergy servers with Intel Xeon 5500 series processors were used to host 14 application server VMs and one database server VM.  The details about this test are in a benchmark report that was put together by Fujitsu and VMware.  The official SAP benchmark site also has added this new result to the Three-Tier SD Benchmark results.

World record TPC-H performance on VMware vSphere 4

In what has become a regular phenomenon, VMware's vSphere 4
has again proved that it can deliver the highest levels of performance for the
most demanding applications. In a recent benchmarking effort by ParAccel, Inc.,
provider of ParAccel Analytic Database (PADB), vSphere 4 delivered the industry's
first fully-audited TPC-H benchmark result in an x86 based virtual environment.
A cluster of 80 virtual machines running PADB, spread across 40 physical hosts,
broke the existing world records for performance[1] and

Some of the key highlights of this record performance are:

  • Ran 1,316,882 Composite Queries per Hour (QphH) @ 1,000GB, 13% higher than the next best
  • Used 40 HP servers, 37%
    fewer than that used for the second best performance result.
  • Achieved a price/performance of US $0.70 per QphH, 7.7 times better than that of the previous
    performance record holder.
  • Took 16 minutes and 23 seconds to load 1 TB database at a rate
    of 3.7 TBs per hour, a load time 8.7
    times  faster than that of the
    previous performance record holder

This remarkable feat shows that vSphere enables efficient
utilization of hardware resources through consolidation and achieves this
without compromising on performance. Organizations can confidently virtualize the
world’s most demanding applications and still realize significant cost and
energy savings.

The setup used for this record feat was,

40 x HP DL 380 G6 Server, each with

  • 2 Intel Xeon x5560, 2.8GHz processors ( 4 cores
    per socket)
  •  72 GB Memory
  • 8 x 300 GB (10K RPM) internal SAS Disks
  • 2 x HP PCIe Dual Port Gigabit network adapters

VMware, ESXi 4.0 update 1

Virtual Machines:
Leader Node:

1 virtual machine with

  • 4
  • 32 GB memory
  • 4
    x 279 GB virtual disk
  • 5
    x virtual NICs
  • ParAccel
    Standard Linux, 64-bit 
  • ParAccel
    Analytics Database, ver 2.5

Compute Node:
79 virtual machines with

  •  4 vCPU
  •  32 GB memory
  •  4 x 279 GB virtual disk
  •  4 x virtual NICs
  • ParAccel Standard Linux, 64-bit
  • ParAccel Analytics Database, ver 2.5

For further details, read the full disclosure report here: http://www.tpc.org/tpch/results/tpch_perf_results.asp

[1] As of  April 13, 2010: HP BladeSystem c-Class 128P
RAC with Oracle Database 11g Release 2 Enterprise Edt., 1,166,976 QphH@1000GB,
5.42USD per QphH@1000GB, available on September 01, 2009.

[2] As of  April 13, 2010: PRIMERGY RX300 S4 with EXASOL
EXASolution 2.1, 1,018,321 QphH@1000GB, 1.18 USD per QphH@1000GB, available on
August 1, 2008

vCenter 4.0 Database Performance Whitepaper Posted

VMware® vCenter database stores meta data on the state of
a VMware vSphere environment and is a key component of vCenter performance.

We recently ran performance micro-benchmarks on a vCenter
4.0 database with Microsoft SQL Server 2008 and published a Performance white

This paper presents the performance results of the
micro-benchmarks and provides best practices information for configuring a vCenter 4.0 database. The paper also provides
information for sizing the server you use to host the vCenter database based on
these performance results. The examples and performance data presented in this
study are specific to Microsoft SQL Server and the paper assumes that you have
a working knowledge of SQL Server.