After the benchmark about Office SharePoint Server 2007 ( 171,000 concurrent users on a single server ), VMware published a new performance report related to its new vSphere 4.1 virtual infrastructure. This time it’s about SAP. Once again the company measured the performance of the system using a single physical host: a Dell PowerEdge R905 4U rackmount server with 4 AMD Quad-Core Opteron 8384 CPUs and 128GB RAM. vSphere 4.1 started serving a single virtual machine with 2 vCPUs and 16GB vRAM, loaded with Novell SUSE Linux Enterprise Server (SLES) 10 Service Pack 2, SAP ERP 6.0 Enhancement Package 4 and IBM DB2 9.7. The test was repeated with multiple VMs, with 2 and 4 vCPUs each. VMware reports that this configuration can serve up to 1,200 concurrent users with 2 or 4 VMs running a grand total of 8 vCPUs. If the amount of VMs and the amount of vCPUs is doubled than this system can serve up to 2,500 concurrent users: Labels: Benchmarks , SAP , VMware

Read the rest here:
virtualization.info

Now that the SPEC has finally released the first industry standard benchmark for hardware and OS virtualization platforms , customers may believe that there’s no more need for the VMmark proprietary framework that VMware released in July 2007 . VMware has a different opinion and last week announced the public beta of VMmark 2.0. While SPECvirt_sc2010 and VMmark 1.x measure the performance of a single virtualization host, VMmark 2.0 has been designed to benchmark a whole virtual data center. This implies measuring complex operations like manual and automated (or DRS-initated) vMotion, Storage vMotion, as well as virtual machines cloning and deployment. On top of that, VMmark 2.0 also features more resource-intensive workloads, including: a multi-tier OLTP workload (DVD Store 2) consisting of a 4-vCPU database VM and three 2-vCPU webserver VMs driving a bursty load profile a multi-tier social networking workload (OLIO) consisting of a 4-vCPU web server and a 2-vCPU database server a mail server workload (Microsoft Exchange2007) consisting of a 4-vCPU mailserver a standby server with just one vCPU For the large majority of customers, benchmarking entire virtual infrastructures may sound completely useless at today. There are even doubts that the market really needs to benchmark a single virtualization host. But going forward, the design of VMmark 2.0 seems a good start to benchmark fully automated Infrastructure-as-a-Service (IaaS) cloud computing platform, which already seems a concrete need looking at the efforts of companies like CloudHarmony , Compuware and even Web Hosting Talk . Of course, until VMware drops its restrictions on VMmark’s EULA, nobody will be really able to use it. Labels: Benchmarks , VMware

Go here to see the original:
virtualization.info

The non-profit Standard Performance Evaluation Corp. (SPEC) has released SPECvirt_sc2010, the first vendor-neutral benchmark to measure the performance of datacenter servers used for virtualized server consolidation. The new benchmark also includes options for measuring power consumption and power/performance relationships. [[ This is a content summary only. Visit my website for full links, other content, and more! ]]

Go here to read the rest:
Virtualization.com

A few weeks ago VMware acknowledged a bug in ESX that translates into poor performance when it runs Microsoft Terminal Services workload on Intel 5500 (codename Nehalem) CPUs with Hyper-Threading (HT) enabled. The issue emerged in independent benchmarks published with the Project Virtual Reality Check (VRC) framework and Citrix has been quick in suggesting that this demonstrates the flaws of the VMware EULA . The founders of Project VRC rerun the benchmark with the patch and the results are considerably better .

Link:
virtualisation.info

Yesterday Cisco published the first VMware VMmark benchmark obtained with its Unified Computing System blade platform B250 M2 and VMware vSphere 4.0 Update 1. The B250 M2 machine, powered by the just released Intel Xeon quad-core X5680 CPUs (codename Westmere) at 3.33GHz and 192GB RAM, scored 35.83 with 26 tiles , a 42% increase over the previous best result obtained by Fujitsu with the RX300 S5 and VMware vSphere 4.0: 25.16 with 17 tiles. The full configuration of this B250 M2 is described here . CRN reports that new UCS systems with the impressive Intel Xeon 5600 CPUs , used for this benchmark, will be available in April.

In July 2008, the VMware Board of Directors voted to remove the founder Diane Greene as CEO of the company. Greene was offered another position that she declined, leaving the company that she created and led through one of the most impressive IPO in the IT history . Two months after her departure, his husband Mendel Rosenblum, left too . Rosenblum co-founded VMware and was the Chief Scientist declining the company vision. The board immediately replaced her with Paul Maritz, a long-time Microsoft executive that joined the EMC ecosystem after his startup Pi was acquired in February 2008.

Here is the original post:
virtualisation.info