Avastu Blog: Sustainable Global Clouds

VMware’s Virtual Center management platform is also mature and straightforward in how an administrator can use it to control resident VMs on a VMware host. VMware’s Virtual Infrastructure Client (VIC) is the administrative user interface to the VMware Virtual Center platform. Microsoft’s System Center-Virtual Machine Manager (SC-VMM) 2008 (we tested a very late beta version which Microsoft guaranteed was feature complete) works with very strong ties to the underlying Active Directory and has an interface that fits right into Microsoft’s System Center scheme, so administrators won’t have to work hard to understand how it works. That said, things from standard management tasks such as viewing simple settings for a VM host to much touted advances features like the ability to migrate ESX VMs to Hyper-V caused SC-VMM to crash repeatedly during testing. Click to see: Microsoft, with its System Center Virtual Machine Manager 2008 software, provides a centralized console for ...

"AMD has an extensive track record of helping enterprises face demanding challenges by enabling flexible, virtualized systems that address major issues including server consolidation, business continuity and disaster recovery," said Patrick Patla, general manager, Server and Workstation Business, AMD. "AMD's latest benchmark result underscores the fact that Quad-Core AMD Opteron processors are more than just four cores and provide innovations such as AMD Virtualization with Rapid Virtualization Indexing to help deliver an ideal platform for addressing real-world data center needs." Rapid Virtualization Indexing (RVI) technology is designed to offer near-native performance of virtualized applications while enabling fast switching between virtual machines (VMs.) The VMmark 1.1 consolidation benchmark, released by VMware in May 2008, ...

Get it here

See it all here and they mentioned also how they got those high figures: Tuning for Performance You might be wondering what parameters we tuned to obtain this kind of performance. The answer will surprise most: we tuned only three parameters to obtain 100,000+ IOPS. We increased the VMFS3 max heap size from 16MB to 64MB (KB article # 1004424). We changed the storage processor’s cache high/low watermark from 80/60 to 40/20. This was done to write the dirty pages in storage cache more often so that Iometer write operations do not wait for free memory buffers. We increased the guest queue length to 100 to make sure that the guest was capable of queuing all the I/O accesses generated by Iometer to the test disks. Don't go on benchmarking your production, or wait!... actually do it and post your results!

Our analysis shows the capacity of the physical machine was 35 users. The virtual machine had a capacity of 30 users - a decrease of 14% from the baseline capacity. This analysis, along with a variety of charts is produced automatically in Load Tester's Load Test Report. Consult the User Capacity section of the test reports for more details on this analysis (see references). The charts shown below are extracted from that report section. The red line on the chart depicts the maximum of each average page duration as measured during each sample period. The location where that line crosses the duration threshold indicates the point where performance no longer met the criteria. Had any errors occurred during the test, the yellow line would depict the error rate and indicate when it crossed the error threshold. Physical Machine Page durations Virtual Machine Page durations Read the whole report here and ask yourself why aren't you doing DaaS.

This is where those tests were carried out: When we were choosing computer models, we set out to choose not the fastest, latest models, but ones that would be a good representation of what most people may have. Certainly, the faster models of these computers will perform even better. Similarly, we had a greater focus on XP simply because it's more prevalent at this point, but we did want to get an understanding of how Vista performed as well. The baseline PC we used was a brand new Fujitsu Lifebook A6025, with an Intel Core Duo running at 1.86 GHz, 1GB RAM, running Windows XP SP2. We chose three Mac models to compare alongside a name brand PC: a MacBook, a MacBook Pro, and a Mac Pro. The MacBook was a 2GB RAM machine, running a 1.83 GHz Core Duo processor. The MacBook Pro was a 4GB RAM machine, running a 2.16 GHz Core 2 Duo processor. And, the Mac Pro was a 4GB RAM machine, running a Quad Core configuration with two 2.66 GHz Dual-Core Intel Xeon processors. Test Types T...

The Xen Project Mission Statement is - Build the industry standard open source hypervisor - Core "engine" that is incorporated into multiple vendors' products • Maintain our industry-leading performance - Be first to exploit new hardware acceleration features - Help OS vendors paravirtualize their OSes • Maintain our reputation for stability and quality - Security must now be paramount • Support multiple CPU types; big and small systems - From server to client to mobile • Foster innovation - Be a great platform for research and experimentation • Drive interoperability - Between Xen-based products - With other virtualization products Check out the Citrix's site

This is my first post on ITtoolbox and an excerpt is here: We all noticed how Oracle and Sun, in the shadows of Oracle's OOW event, announced their virtualization platforms. Both of these are based on Xen. Oracle came with its own story of how their VM's were 3 times faster than any other virtualization vendor. They also claimed that "the vendors who claim to have the only live migration on the planet" should stop claiming that as it too is becoming a commodity. So why did VMware start talking about Microsoft suddenly? Simple, because Microsoft is readying its "bare metal" hypervisor and is planning a major strike on claiming back its "children" (read guest OSs) to run back to their "biological mom" (read MS hypervisor). Microsoft's base is massive and VMware is building its fortunes on its OS stack, Application stack (SQL Server, Exchange). Other bigger, mother-of-all apps like Sharepoint, that Steve Ballmer touted as the "next ...

and they have a lot of itneresting benchmarks. Go here.

The tests were performed on servers configured with 2, 4, 6 and 8 gigabytes of main memory at various transaction processing load levels. The results show that for certain configurations and at certain load levels the Intel Xeon based server was 2.4 to 11.7 percent more power efficient while in other cases the AMD Opteron based server was 9.2 to 23.1 percent more power efficient. In addition, when the systems were idle and waiting for transactions to process, the AMD server was 30.4 to 53.1 percent more power efficient. Power consumption while the servers are idle is particularly significant since many servers spend most of their time waiting for work. A November 16, 2006 press release(1) from IBM quotes a report by the Robert Frances Group(2) which states that on average servers in datacenters are idle 80 to 85 percent of the time. Other observations that can be made from the test results include: 1) Larger memory configurations deliver both higher throughput and better power efficien...

CNET recently did a performance comparison between Parallels Desktop and VMware Fusion. Surprisingly enough, their benchmark also suffered from unrealistic scenarios and vendor bias. Pretty much everything about this test is upside down: * It does not make sense to use an exclusive monster 8-core desktop for benchmarking Windows on Mac. Most people use laptops with 2 CPU cores and 2GB of RAM at most. * Since Mac OS is the primary OS, it does not make sense to give both cores to a Windows VM that runs Word, Excel and Outlook. Office applications don't benefit from multiple CPUs. Perhaps that's why default configuration of Fusion is a single-CPU VM. Come on – most of the VMmark (server benchmark) workloads run in a single-core VM. Yet, desktop benchmark is run with dual cores – does not make much sense to me. * It does not make sense to use Vista inside a VM. Most people run XP because Vista license only allows the most expensive Vista SKUs to run inside VM, not to me...

When a lot of talk is going around the net, it is sometimes also wise to take a backseat and reap the fruits of all those hard working engineers and staff of all the companies and give them the credit they deserve. They could be Microsoft developers working their jingles off making Office a better product; they could be XenSource programmers working hard to develop an innovative solution; they could even be VMware Fusion guys; and then they would be parallels engineers who will make yet another innovative stuff. I truly loved their "coherence". Truly! For the same reason, we see how Fusion excelled against the great product from Parallels. We are sure Parallels will strike back. One of Fusion's primary advantages in the test is that it takes advantage of two CPU cores, while Parallels only takes advantage of one. We recently published an in-depth look at Fusion along with commentary from Pat Lee, product manager of at VMware. Lee said that VMware generates "less ove...

In order to produce the benchmark, VMware has avoided the relatively easy approach of measuring the performance of virtual applications software running on specific machines. Instead, the benchmark measures the scalability of heterogeneous virtualised workloads. According to VMware: "It provides a consistent methodology so benchmark results can be compared across different virtualisation platforms." The result is that, by using the benchmark, companies should be able to "make appropriate hardware choices, and compare the performance and scalability of different virtualisation platforms", VMware says. Nice stuff this benchmarking. But will the clients really benchmark their stuff. Or lwetm e put it more bluntly, how the heck are you going to tell a client/customer to benchmark if they just have no clue of why a benchmark is done in the first place. Now that will not mean that all the clients are that way. I am pretty sure that besides silencing folks who have run tes...

Q: Why does your EULA prohibit the publishing of benchmark results? A: We endorse benchmarking but we want people to follow proper procedures. There's not much known about benchmarking VMs. For example, how do you make an industry standard benchmark map onto virtualisation? Take TPCC -- you modify it, run multiple workloads, how do you get meaningful numbers? Our EULA provides guidance on how to do this and want to help people get the right numbers. We want people to use VMmark and learn how to benchmark virtual systems. Read the rest here .

Read for instructions and to download go here here .

Chris Wolf published some Virtual Iron's benchmark results and commented that many of the people believe XenSource's benchmark to be more accurate. How did they come to the conclusion, I don't know. So overall the Windows Server 2003 VM was able to perform at or below a 3% performance degradation. The Virtual Iron tests followed the same benchmark pattern used by VMware. If you would like to see the VMware results and also get more detail on what each individual benchmark is testing, take a look at VMware’s document “A Performance Comparison of Hypervisors.” Keep in mind that the Xen performance numbers in the VMware paper are under significant debate, with most of us (myself included) seeing Simon Crosby’s Xen benchmark numbers as being more accurate. Read more...

These results show that ESX Server not only achieves higher throughput than VMware Server for a single VMmark tile (6 workload VMs) but also exhibits better scalability when a second tile is added. This behavior is a natural consequence of the different virtualization approaches taken by the two products. VMware Server runs on top of a heavyweight, general-purpose host operating system, which manages the hardware resources. Read the article.

XenSource's 3.2 Enterprise Version beta was compared with the ESX 3.0.1. Our performance comparison of XenSource XenEnterprise 3.2 beta to VMware ESX Server 3.0.1 is now available, having been vetted and approved by VMware’s benchmark police. As hinted at earlier, we have nothing to be ashamed of, and lots to be proud of, given the fact that XenEnterprise hasn’t reached its first birthday yet. In most tests, the results for the two virtualization platforms are essentially indistinguishable; in the Linux test, the value of paravirtualization is borne out strongly. Get the PDF .

Bruce Herndon reports on VMware blog: One of the many benefits of having a reliable and representative benchmark like VMmark is that it allows vendors to demonstrate and compare their platforms when running virtualized environments. Get the PDF!

Bruce Herndon tries out the VMmark on new hardware. These two different disk configurations highlight some interesting tradeoffs and tuning opportunities exposed by VMmark. The single LUN configuration utilizing six disks has the benefit of providing high disk throughput for one VM at the expense of scalability if multiple disk-intensive VMs are running. On the other hand, creating multiple LUNs provides both good predictability and excellent scaling but limits the total throughput of any single VM by providing only a subset of the hardware resources to each one. From a benchmarking perspective, the multi-LUN approach is clearly better since it results in a higher overall score. Honestly, we have reported several benchmarks and tools being used were either beta or some of them reported results that really didn't matter. VMware is also a member of the SPEC committee but still results like these speak volumes. Benchmarking is not about VMware is better than Parallels is better that X...