Virtualization Report | David Marshall | April 2006

MORE ENTRIES

April 2006

April 30, 2006 | Comments: (0)

Sun Releases a Consolidation Tool for Sun Fire Servers V1.0

Sun has recently launched a powerful tool to help with the process of server consolidation on Sun Fire T1000 and T2000 platforms. The functionality is achieved using a simple and easy to use interface that offers context sensitive help. To make things easier, configuration details are saved so that the configuration can be reused easily on other Sun Fire T1000 or T2000 systems. Even more interesting, the tool is being made available as an open source application.

Quoting from Sun:

The Consolidation Tool for Sun Fire Servers V1.0, Sun Fire T1000 and T2000 Edition, is a software tool designed to simplify the task of consolidating multiple applications onto the Sun Fire T1000 or T2000 platform. In particular, this software allows the easy definition and creation of Solaris Containers. The tool deploys applications into processor sets where appropriate and allocates CPUs into processor sets in a way that optimizes performance. The software also takes full advantage of Solaris Containers (including Solaris Zones partitioning technology, pools, psets, and scheduling classes), making intelligent choices between full-root and sparse zones.

The tool is designed to offer the following features:

Captures customer's workload consolidation requirements (no assumed knowledge of the Solaris OS)
Builds resource pools, psets, and Solaris Zones (complies with best practices for Sun Fire T1000 and T2000 servers)
Deploys a short list of public domain applications into Solaris Zones (automatically installs nominated apps)
Provides a report summarizing the deployment (executive summary in html or possibly pdf)

The tool can be run on any of the following operating systems:

Solaris OS on SPARC platforms
Solaris OS on x64/x86 platforms
Linux
Mac OS X

To download the tool or a presentation outlining the tool and its features, click here.

Posted by David Marshall on April 30, 2006 03:22 PM

April 29, 2006 | Comments: (0)

InfoWorld Virtualization Report Podcast

Today we talk about virtualization of the desktop with two different approaches. Have you heard about VMware's Virtual Desktop Infrastructure Alliance? Pull up a chair. And then... Intel gets into the game with its latest offering, vPro. listen LISTEN!

Posted by David Marshall on April 29, 2006 04:49 PM

April 29, 2006 | Comments: (0)

Virtual Server 2005 R2 is getting a service pack

It looks like Microsoft's Virtual Server 2005 R2 product is about to get its very own service pack. Currently in beta, the Virtual Server 2005 R2 SP1 is now available for download. The free service pack will add support for hardware-assisted virtualization leveraging the capabilities developed by AMD and Intel, improved backup services, and more. By supporting both AMD Virtualization and Intel Virtualization Technology, the product will offer improved performance for non-Windows guest operating systems and provide better interoperability.

Virtual Server 2005 R2 Service Pack 1 Beta 1, available now, includes:

Support for Intel Virtualization Technology
Virtual Server's host clustering step-by-step guide

Virtual Server 2005 R2 Service Pack 1 Beta 2, scheduled for Q4, includes Beta 1 features plus:

Support for AMD Virtualization
Volume Shadow Service support
Active Directory integration
and Expanded OS support

Beta 1 download is available with a Microsoft Connect account. Login and select the Virtual Server 2005 R2 SP1 Beta program from the list of available programs that appear here.

Posted by David Marshall on April 29, 2006 04:37 PM

April 29, 2006 | Comments: (0)

Intel's vPro - Virtualization for the business desktop

Intel Corporation made an announcement about a new platform that it intends to launch, vPro. Expected sometime in the second or third quarter of this year, Intel hopes it will change the way desktops function and the way they are managed within a distributed business environment.

The vPro platform is powered by Intel's Conroe dual-core processor and backed by a 64-bit micro architecture that has built-in power and energy saving features and includes the second generation of Intel Active Management Technology (Intel AMT) along with Intel's Virtualization Technology (Intel VT). When combining Intel AMT in the chipset and Intel VT in the dual-core processor with industry enabled software solutions, tomorrow's business desktop PC will offer businesses and IT staff superior manageability and improved security.

Intel's chief Paul Otellini has been quoted as saying "It's time to reinvent the desktop". With Intel AMT, businesses gain assistance with inventory and management, and diagnosing and repairing PCs even when the system is turned off or has crashed. With the second generation Intel AMT technology, PCs that become infected can quickly be isolated from the network environment and then notify IT staff when the threat is removed.

Intel VT goes one step further, it helps strengthen PC security by creating a separate independent hardware-based environment inside of the PC allowing IT administrators to create a dedicated and tamper proof service environment or partition, where tasks can run independently and isolated from the main operating system as well as from the end user.

Support for Intel vPro technology will be available over the next year from companies such as Adobe, Altiris, Check Point, Cisco, Computer Associates, LANDesk, Microsoft, and Symantec to name a few.

It might not be long until your business PC sports the new vPro logo!

Posted by David Marshall on April 29, 2006 09:44 AM

April 26, 2006 | Comments: (0)

InfoWorld Virtualization Report Podcast

Today we talk about Intel powered Mac systems, dual boot or virtualize? And then... Microsoft enters virtualization management space with a new MOM pack and project Carmine listen LISTEN!

Posted by David Marshall on April 26, 2006 08:58 PM

April 26, 2006 | Comments: (0)

Egenera Granted Three New Patents

The industry throws around a lot of buzz words - one of which is "utility computing", and it is becoming a more commonly talked about computing solution. Egenera's Bladeframe System seems to fit in this space quite nicely. It combines its stateless servers with a software that virtualizes processing, storage, and networking resources into a "computing fabric" allowing IT shops to rapidly provision systems.

To help maintain their positioning in the industry, Egenera continues to innovate, and the engineering team has recently been granted three new patents to help their leadership position.

Quoting from their press release:

Egenera's unique BIOS technology virtualizes the operations provided by standard server BIOS, allowing those I/O operations to be performed to devices that are not physically attached to the server. Specifically, the virtual BIOS accesses virtual console and virtual storage resources necessary to transform diskless, stateless Egenera® Processing BladeTM modules into virtual servers. This technology is a key enabler for virtualizing server configuration while providing the standard BIOS interfaces required by enterprise server operating systems.
The newly patented interconnect technology, referred to as the BladeMateTM system, integrates high-speed fabric connections, blade management and power distribution in a hot-plug connection scheme. Multi-level alignment between mating connector components ensures highly reliable operation. This embedded connection technology allows Egenera to deliver blades based on virtually any processor architecture.

Working in concert with BladeMate is the newly patented BladeLatchTM system, a handle mechanism for easily installing, securing and removing blades from the Egenera BladeFrame® chassis. Together, these technologies enable Egenera customers to leverage advances in processor technology by upgrading blades without disrupting overall datacenter operations.

To find out more about the company and their products, visit their Web site, here.

Posted by David Marshall on April 26, 2006 04:43 AM

April 25, 2006 | Comments: (0)

Microsoft Fills a Management Void but with VMware on its Heels

If you’ve been using Microsoft Virtual Server 2005 for any period of time in an enterprise environment, it is safe to say, you have probably experienced the lack of management tools. In an effort to play catch up with VMware’s management solutions, Microsoft has recently announced two products to hopefully fill that void.

The first product, Microsoft Virtual Server 2005 R2 Management Pack, is a management solution that works with Microsoft Operations Manager or MOM. It enables you to monitor the physical server host (running either Microsoft Virtual Server 2005 or Microsoft Virtual Server 2005 R2) as well as the host server’s virtual machines. The management pack provides monitoring and reporting for the following scenarios:

Virtual Server
Monitors Virtual Server service availability
Detects configuration errors
Detects critical error conditions
Provides a graphical mapping of virtual machines to their virtual machine hosts

Virtual Machines
Monitors virtual machine health state
Detects start and restoration failures
Detects save failures
Detects critical error conditions
Monitors processor, memory, and disk utilization

Reporting
Identifies good candidates for conversion to virtual machines
Views performance history of a computer over a specified period of time
Views a summary of host servers and virtual machines
Views CPU, memory, and disk utilization for a virtual machine

The second product announced was Microsoft’s Virtual Server Manager, codenamed Carmine, which is expected to enable administrators to add, move, and manage virtual machines on its virtualization platform. Hopefully, one of its key features will help with the huge problem of patch management, allowing the administrator to move the virtual machines from one host to another while the host server is patched and rebooted before moving the virtual machines back. While the product is expected to have some of the basic features of VMware’s VirtualCenter product, it certainly will not be as sophisticated, as VMware’s product has been in the hands of its users for two and half years now.

But is it going to be enough? VMware is still on target to release its next generation platform, VMware ESX Server 3.0 and VMware VirtualCenter 2.0 by the end of this quarter. Adding on the new Distributed Resource Manager (DRM) and Distributed Availability Service (DAS) features to their product, along with other new management and feature functionality, seems to keep them a few steps ahead in the race yet again. VMware is determined to hold that number one spot it earned so many years ago.

Posted by David Marshall on April 25, 2006 04:59 AM

April 22, 2006 | Comments: (0)

KernelTrap Interviews Andrey Savochkin - A Lead Developer on OpenVZ

KernelTrap.org offers an interesting interview with Andrey Savochkin, a lead developer of the kernel portion of OpenVZ, an operating system-level server virtualization solution. In the interview, Andrey goes into great detail about what virtualization is and how it works, the differences between hardware-level and operating system-level virtualization, and how OpenVZ compares to VServer, Xen, and User Mode Linux. Andrey is now focused on getting OpenVZ merged into the mainline Linux kernel because he believes virtualization is the next big step, "comparable with the step between single-user and multi-user systems."

Jeremy Andrews: OpenVZ is described as an "Operating System-level server virtualization solution". What does this mean?
Andrey Savochkin: First, it is a virtualization solution, that is, it enables multiple environments (compartments) on a single physical server, and each environment looks like and provides the same functionality as a dedicated server. We call these environments Virtual Private Servers (VPSs), or Virtual Environments (VEs). VPSs on a single physical server are isolated from each other, and also they are isolated from the physical hardware. Isolation from the hardware allows to implement on top of OpenVZ an automated migration of VPSs between servers that does not require any reconfiguration for running the VPSs on a very different hardware. A fair and efficient resource management mechanism is also included, as one of the most important components for a virtualization solution.

Second, OpenVZ is an operating system-level solution, virtualizing access to the operating system, not to the hardware. There are many well-known hardware-level virtualization solutions, but operating system-level virtualization architecture gives many advantages over them. OpenVZ has better performance in some areas, considerably better scalability and VPS density, and provides unique management options in comparison with hardware-level virtualization solutions.

...

Jeremy Andrews: How does OpenVZ improve upon other virtualization projects, such as VServer?

Andrey Savochkin: First of all, OpenVZ is a completely different project than VServer and has different code base.

OpenVZ has bigger feature set (including, for example, netfilter support inside VPSs) and significantly better isolation, Denial-of-Service protection and general reliability. Better isolation and DoS protection comes from OpenVZ resource management system, which includes hierarchical CPU scheduler and User Beancounter patch to control the usage of memory and internal kernel objects. Also, we've invested a lot of efforts in the creation of the system of quality assurance, and now we have people who manually test OpenVZ as well as a large automated testing system.

Virtuozzo, a virtualization solution built on the same core as OpenVZ, provides much more features, has better performance characteristics and includes many additional management capabilities and tools.

Jeremy Andrews: What are some examples of hardware-level virtualization solutions?

Andrey Savochkin: VMware, Xen, User Mode Linux.

Jeremy Andrews: How does OpenVZ compare to Xen?

Andrey Savochkin: OpenVZ has certain advantages over Xen.

OpenVZ allows to utilize system resources such as memory and disk space much more efficiently, and because of that has better performance on memory-critical workloads. OpenVZ does not run separate kernel in each VPS and saves memory on kernel internal data. However, even bigger efficiency of OpenVZ comes from dynamic resource allocation. Using Xen, you need to specify in advance the amount of memory for each virtual machine and create disk device and filesystem for it, and your abilities to change settings later on the fly are very limited. When running multiple VPSs, at each moment some VPSs are handling load burst and are busy, some are less busy and some are idle, hence the dynamic assignment of resources in OpenVZ can significantly improve the utilization of resources. With Xen, you have to slice the server for the worst-case scenario and maximal resource usage by each VPS; with OpenVZ you usually can slice basing on average usages.

OpenVZ provides more management capabilities and management tools. To start, OpenVZ has from out of the box ability to immediately create VPSs based on various Linux distributions, without preparation of disk images, installing hundreds of packages and so on. But most importantly, OpenVZ has the ability to access files and start from the host system programs inside VPS. It means that a damaged VPS (having lost network access or unbootable) can be easily repaired from the host system, and that a lot of operations related to management, configuring or software upgrade inside VPSs can be easily scripted and executed from the host system. In short, managing Xen virtual machines is like managing separate servers, but managing a group of VPSs on one computer is more like managing a single multi-user server.

Operating system inside Xen virtual machine is not necessarily able to use all capabilities of the hardware; for instance, support of SMP and more that 4GB of RAM inside virtual machines will appear only in Xen 3.0. OpenVZ is as scalable as Linux when hardware capabilities increase. SMP and more than 4GB have been supported in OpenVZ from the very beginning. Recently we've built OpenVZ for x86_64 platform, and it was a straightforward job not requiring going into architecture details. So, OpenVZ is far more hardware independent than Xen, and hence is able to start to use new hardware capabilities much faster.

There is one point where Xen will have certain advantage over OpenVZ. In version 3.0, Xen is going to allow to run Windows virtual machines on Linux host system (but it isn't possible in the stable branch of Xen).

Again, I need to note that the above describes my opinion about the main differences between OpenVZ and Xen. Virtuozzo has many additions to OpenVZ, and, for instance, there is Virtuozzo for Windows solution.

Jeremy Andrews: How does OpenVZ compare to User Mode Linux?

Andrey Savochkin:
What I've said before about advantages of OpenVZ over Xen also apply when OpenVZ is compared with User Mode Linux.

The unique feature of User Mode Linux is that you can run it under standard debuggers for studying Linux kernel in depth. In other aspects, User Mode Linux does not have as many features as Xen, and Xen is superior in performance and stability.

Read the entire interview, here.

Posted by David Marshall on April 22, 2006 08:40 AM

April 22, 2006 | Comments: (0)

Have You Met Virtuozzo?

If you are new to server virtualization, you have probably been bombarded with literature and information about products such as VMware GSX Server, VMware Server, VMware ESX Server and Microsoft Virtual Server. True, these are all fantastic products and do a great job accomplishing exactly what they set out to do, but are they the only solutions out there? The short answer is no. While these products have been around for years in one form or another, they aren't alone and the space has been heating up for quite some time producing more and more entries into the market.

One such product is Virtuozzo from SWSoft. And while they take a different approach to server virtualization than say VMware, you might want to take a closer look at them. Paul Venezia does just that in his InfoWorld article, "Virtuozzo commands virtual server stage". In it, he writes:

If you think of server virtualization in terms of VMware, SWSoft Virtuozzo will turn that thinking around.
Whereas VMware virtualizes and emulates the hardware layer, Virtuozzo runs with the native OS on the host server and creates VPSes (virtual private servers), by generating chroot jails for the host. It then tightly controls the network and kernel layers to deliver packets correctly and to handle disk, memory, and CPU quotas.

I looked at Version 3.0 of SWSoft’s Virtuozzo for Linux. First released in 2001, Virtuozzo for Linux has come a long way, and this version has a lot to offer, not the least of which is the v2.6 kernel base -- better late than never.

SWSoft’s approach to server virtualization has some definite benefits over VMWare’s. Among them, it requires much less overhead, so more virtual servers can be run from a single server without noticeable performance degradation. Further, the management of the servers is simpler.

...

The templates Virtuozzo provides are broad in range, though by no means complete. Fortunately, SWSoft has provided plenty of tools for admins to roll their own templates, which is generally as simple as providing a pointer to a standard RPM (Red Hat Package Manager) for any given distribution, or simply a directory of files. After the template is created, it can be applied to one or more VPSes in a single command.

Also present in the management console are control panels for quickly modifying Apache and Sendmail configurations on a per-VPS basis, and monitoring panels for each VPS, complete with Windows Performance Monitor-style live-utilization graphing. Very cool.

Paul gives SWSoft Virtuozzo for Linux 3.0 an excellent rating of 9.3. To read more of his review and findings, check out his article, here.

And to find out more about SWSoft Virtuozzo, check out their Web site, here.

Posted by David Marshall on April 22, 2006 07:08 AM

April 22, 2006 | Comments: (0)

IBM Launches New System x Servers and Software Targeting Large Scale x86 Virtualization

IBM made an announcement that it would address unprecedented demand for large scale x86 virtualization technologies with its System x scaleable servers featuring IBM's Enterprise X-Architecture technology, X3, as well as new software to help speed virtualization efficiencies. System x comes in response to the large growth of virtualized systems. And IBM announced three high performance systems designed to deliver the greater virtualization, utilization, and performance focus that is needed in today's market.

Quoting from IBM's official press announcment:

"IBM has the hardware, software and services capabilities required to provide enterprise class virtualized solutions to our most demanding clients," said Susan Whitney, general manager, IBM System x. "IBM's strong partnership with industry leaders like VMware as well as its own Virtualization Engine technologies that are built into System x, allow us to deliver the innovative solutions needed to help customers drive efficiencies in their business operations."
...

Based on IBM's Virtualization Engine technologies, IBM also unveiled breakthrough software that recognizes untapped utilization and hidden servers across clients' networks and helps identify opportunities to consolidate and virtualize x86 systems. More than 600 IBM Business Partners have been certified to use the IBM Consolidation Discovery and Analysis Tool or CDAT. Tools such as CDAT have enabled IBM Business Partners to accelerate virtualization adoption with more than 65 percent of our virtualization client engagements coming from business partners. Combined with IBM Director, the new software creates a powerful portfolio of datacenter automation and virtualization tools for the x86 segment.

Traditionally each x86 server runs only one application and therefore can have a low utilization rate such as 8 to 12 percent. By pooling, managing and optimizing IT resources across servers, virtualized systems can improve the economics and operations of under-utilized IT assets. The CDAT software has helped hundreds of customers identify and improve performance and in cases even quadruple a client's computing utilization.

...

Pricing and Availability
IBM's new systems for the high-performance enterprise computing segment of System x will be available in May 2006, for virtualization, database, ERP and CRM applications.

System x3950 -- Featuring high-performance capabilities, flexible partitioning and mission-critical availability, ideal for database, ERP and server consolidation; starting at US $17,499. (Formerly x460)

System x3850 -- Compact four-processor server for 64-bit and mid-tier software applications; starting at US $6,999. (Formerly x366)

System x3800 -- High-performance, multi-purpose server with large internal storage capacity, ideal for remote and branch-office solutions; starting at US $4,999. (Formerly x260)

Working together, System x and IBM storage can help customers create a virtualized environment with management across the entire IT infrastructure, reducing the operational risk of implementing complex solutions. System x servers are supported by the broad range of cost effective IBM disk, tape, software, and storage networking solutions.

Read the entire IBM press release, here.

Posted by David Marshall on April 22, 2006 06:23 AM

April 20, 2006 | Comments: (0)

Parallels Continues to Impress

Say what you want, but Parallels is not a company that is just sitting on the sidelines watching the cars go by.

Their Mac OS X Workstation product has now reached Beta 4 status. Yes, if you are keeping score at home, that was awfully quick. But here it is. With over 70,000 people beta testing the product, Parallels has been able to enhance the offering with the following:

Support for full screen mode
Dual-monitor support
Improved real CD/DVD support
Improved, easier to use interface
Better sound support, including a fix for the Windows 2000 guest OS
Networking kernel panic fixed
Bridged networking issues fixed
WiFi networking issues fixed
Fixes for some of the issues encountered when entering "Sleep" mode when a VM is running
And many other minor bug fixes

But the company is not happy with simply getting their Workstation product to market. No, in fact, they are now offering a free beta release of a product called Parallels VM Compactor 1.0, a powerful and easy to use virtual hard disk management tool that optimizes performance of any Parallels, VMware, or Microsoft virtual machine running a Windows 2000, 2003, or XP guest operating system by compacting its virtual hard disk by up to 80%.

By compacting the virtual hard disk:

it effectively results in better performance and faster disk access
it reduces the size of expanding virtual hard disks to more effectively use real hard disk space
and it creates smaller virtual machine images that are easier to store, move, and share.

If you want to give Parallels VM Compactor a try, you can download it here.

Posted by David Marshall on April 20, 2006 08:13 PM

April 19, 2006 | Comments: (0)

Neterion Announces Hyperframe, a 10 GbE I/O Virtualization (IOV) Architecture

As virtualization expands into hardware, IOV adds a key element to enable end-to-end virtualization in the data center.

Neterion, Inc. announces that the company has developed Hyperframe, an I/O Virtualization (IOV) architecture that expands hardware support to I/O devices. This technology will span the data center and provide added benefits from servers to storage devices. Blade server environments will be particularly affected.

Quoting from Neterion's official press release:

"Today server and storage virtualization is delivering significant cost reduction and asset optimization benefits to IT managers in corporate data centers," said Dave Zabrowski, President and CEO of Neterion. "Hyperframe is designed to provide hardware support at the I/O component level. Neterion's line of products is offering exclusive features that allow optimization of the network interface in a virtualized environment."
Features included in the Hyperframe architecture are: multiple separate receive and transmit queues, independent Direct Memory Access (DMA) engines, separate network addresses, ability to classify and steer receive traffic, sophisticated interrupt schemes, separate register sets, separate copies of the configuration space, ability to direct packets through the PCI-Express fabric, etc.

...

HP and Neterion are among a select group of companies actively involved in virtualization standardization in the PCI-SIG IOV Working Group. The group is charged with developing a specification that extends the PCI-Express protocol to natively support virtualization in compute client, server and storage environments.

"NextIO has focused on developing PCI Express-based shared I/O solutions since its inception," noted KC Murphy, CEO of NextIO, another key member of the IOV Working Group. "We're very pleased to work with Neterion as our technologies are complementary to each other. The combination of Hyperframe with our shared I/O model allows the development of flexible and versatile IOV devices, enabling extremely agile server architectures. Implementing shared I/O in blade servers, for example, will lower TCO, future-proof blade chassis and simplify management for IT managers."

Benefits to end users of an IOV-based networking architecture include:

Cost savings Replacing multiple 1 GbE interfaces with a single 10 GbE adapter reduces complexity and administration costs
Flexibility Dynamically allocating bandwidth across system components rather then limiting bandwidth to a fixed amount per component
Performance Relieving software from classification and routing of frames resulting in significant reduction of CPU utilization
Isolation Every OS is presented with its own independent I/O path and is unaware of shared resources with other OS images
Reliability Users can implement redundancy and fail-over capabilities with the addition of a second 10 GbE interface that results in dynamic management and virtualization of bandwidth

Neterion has developed a whitepaper that provides technical details on the Hyperframe IOV architecture. It is available for download at: http://www.neterion.com/solutions/applications.html.

Posted by David Marshall on April 19, 2006 04:40 PM

April 19, 2006 | Comments: (0)

Storage Subsystem Performance in VMware ESX Server

If you've ever used VMware ESX Server, you probably wondered which virtual SCSI adapter is the right choice. You may have asked, do I accept the default choice that goes along with my guest operating system? Or does the newer virtual LSI Logic adapter offer me better performance than the older virtual BusLogic adapter? These are good and valid questions to ask. When creating a virtual machine, you definitely want to try and create your guest operating system to perform at its peak potential. So what is the answer? You might be interested to know that VMware has released a white paper discussing this very question. In it, they write:

The storage subsystem is a critical determinant of system performance. The key to good storage performance is to identify factors and system configuration settings that affect performance and understand how to set these in order to achieve the best results.
The basic determinants of performance are the operating system, the data transfer size, and the access pattern. In the virtual machine environment, the drivers for the available virtual adapters are also a factor. VMware ESX Server virtual machines can use virtual BusLogic and virtual LSI Logic SCSI adapters. The default driver for a virtual machine depends on the guest operating system. For example, Windows 2000 guests use the Microsoft-supplied BusLogic adapter by default, while Windows Server 2003 guests use the LSI Logic adapter by default.

This document provides a characterization of storage performance for a VMware ESX Server system with an EMC CX500 SAN as the storage back end. The goal is to provide performance data and system resource utilization at various load levels. Throughtput, I/O rate, and response time for various data sizes and access patterns provide sizing guidelines. This baseline data is expected to help debug performance problems and facilitate server consolidation for I/O intensive workloads.

To read how they performed the testing, what performance results and metrics were achieved, and what conclusions were drawn, download the white paper, here.

Posted by David Marshall on April 19, 2006 02:29 PM

April 19, 2006 | Comments: (0)

Intel VT: Making Virtualization Easier?

I'll admit it - I was one of those people sitting around waiting to upgrade my machine after reading all about how AMD and Intel were going to revolutionize the way virtualization would run on my computer. The early buzz around Pacifica and Vanderpool was overwhelming! So much so that I even started downloading and reading articles and papers from Gerald J. Popek and Robert P. Goldberg (such as their 1974 article "Formal Requirements for Virtualizable Third Generation Architectures")just to try and get a better understanding of what was coming around the corner.

And while I still haven't personally upgraded to this latest virtualization technology, I am still curious and interested to see what it can do, now that it has hit the market. A friend of mine, who is obviously less cheap than I am, recently purchased an Intel VT dual core machine and is running Parallels Workstation on it. Although we didn't run any benchmarking software, it appeared to the naked eye that the virtual machines had been given a slight performance boost. But did VT really do anything of the kind?

According to BCM TechPlanet, perhaps not. They wrote:

...Speaking with Diane M. Bryant, Vice President, Digital Enterprise Group & General Manager, Server Platforms Group at the Intel Developer Forum (IDF) in Taiwan, she said: “VMware fully embraces VT so that’s a good indicator, but performance is something that will evolve over time.”
Essentially, Intel does not promise a performance gain. She explained that VT was developed to add reliability to virtualized systems, and is also a way to simplify the software development code.

She also added that it aids the choice of operating systems...

After reading the marketing and all the research, perhaps I saw what I wanted to see. Or was there really a noticeable difference in performance? I'm sure my next upgrade will include AMD Virtualization or Intel VT, but I think I'll hold out for a processor with more cores.

What are your thoughts? Has anyone done any benchmark testing?

Posted by David Marshall on April 19, 2006 01:46 PM

April 13, 2006 | Comments: (0)

Shipping a Data Center Overseas via Drag and Drop

Have you gone through an audit of your data center? Trying to come up with cost saving ideas and sources of efficiency? Want to get the most out of your underutilized hardware? Are you thinking of virtualizing your data center? That is exactly the problem and solution chosen by many companies in the recent months. The question becomes, how do I go about it? Read about how one company did it, AXA Tech, and how they did it with ease with the help of VMware virtualization and PlateSpin PowerConvert.

Quoting from InformationWeek:

"Armed with a complete picture of the data center landscape, AXA Tech chose PlateSpin PowerConvert to automate the process. PowerConvert would remotely migrate the software layer on their dedicated physical servers in Shibuya to the VMware virtual infrastructure in Melbourne and Shirokane. PlateSpin leverages its patent pending OS Portability technology to decouple the software layer (data, applications, and operating systems) from the hardware and stream it over a network to the destination of choice - physical servers, virtual machines, blade servers, or image archives. With a simple drag-and-drop, AXA Tech was able to remotely stream the data, applications, and operating systems from the dedicated hardware in Shibuya, to multi-use VMware ESX servers in Melbourne and Shirokane. On-the-fly, PlateSpin PowerConvert was also used to remotely right-size the destination servers to match resources with the new workload needs."

Read the entire case study, here.

Posted by David Marshall on April 13, 2006 04:55 AM

April 12, 2006 | Comments: (0)

Boot Camp or Virtualization? Which is it?

Will owners of Mac computers really use Boot Camp? Or is it more of a marketing ploy? I've been reading that there are some people who have experienced some serious problems with this dual boot solution. One such issue, after partitioning their hard drives and installing Windows XP Professional, they can no longer boot back into OS X. Now, call me silly, but most people who go out and purchase a Mac might find this "feature" a little annoying. Again I ask, is Boot Camp really the answer or is it just marketing? If you are using your Mac OS X, and suddenly have the need or urge to run something that requires Microsoft Windows, are you completely satisfied with having to stop what you are doing, shutdown any applications, and then reboot your system to launch Windows XP? Wouldn't it make more sense to leverage a virtualization platform on your OS X installation and simply boot Windows XP inside of a virtual machine without having to shut anything down on the OS X side?

A friend and colleague of mine recently purchased a Mac PowerBook with a brand spanking new Intel Dual Core processor, complete with Intel's VT. As expected, the Mac OS X installation screams! We tested out Boot Camp, and unlike some of the unfortunate people that I read about, we didn't have an issue. Windows XP not only installed just fine, but we were able to cleanly and quickly reboot from one partition and operating system to the next. Being virtualization enthusiasts, we had to try Parallels Workstation 2.1 for the Mac. While we found many of the same bugs and missing features that other beta testers have reported to Parallels, I didn't expect that the product would perform as well as it did for a beta release. Windows XP Professional ran extremely well inside of a virtual machine, and it not only offered more capabilities as a virtual machine, but it also seemed to run much faster than it did in its own partition utilizing Boot Camp.

So while Parallels and other virtualization platforms make their way to the Mac OS, I can only wonder why someone would opt for a dual boot solution rather than take advantage of what virtualization offers. Maybe someone can explain it to me; I very well could be missing something.

Posted by David Marshall on April 12, 2006 05:19 PM

April 12, 2006 | Comments: (0)

Parallels for Mac OS X Beta 3 is ready

If you were one of the 50,000 people to beta test Parallels' Workstation 2.1 for Mac OS X, your constructive feedback may very well have helped shape the future of the product by identifying bugs and issues at an extremely fast pace. The proof of your beta testing efforts is evident with a quick release of Beta 3 of the Workstation product.

You spoke, they listened. Here are some of the fixes and improvements made to the product:

1. No need to reinstall Parallels Workstation if it's moved from the default file location.
2. Mac OS X no longer restarts when Parallels Workstation is left running and host wakes up from "Sleep" mode
3. Kernel no longer panics when working with several VMs
4. Improved speed and performance
5. Idling guest OS now consumes only 1-2% of CPU power
6. Bridged networking issues fixed
7. Sound support introduced
8. Improved wi-fi support, especially for MacBook Pro laptops
9. Improved mouse synchronization tool
10. Keyboard mapping and repeating bugs fixed
11. CD/DVD problems fixed
12. Many other minor bug fixes

In addition, there are a few known issues that are being resolved now:

1. Do NOT suspend your VM when sound is playing inside guest, as it may reset Mac OS X.
2. VMs are created with the Sound device pointing to the Null Device. It's important to point it to "Default Audio" manually to have sound enabled in a VM.
3. Sound may tremble at the beginning of playing
4. USB is not working yet, but will be very soon
5. Video output is still a bit slow
6. Image Tool is not available yet, but will be very soon.

What are your thoughts on how fast Parallels is making progress? Have you used the product? What do you think about it?

Download the new beta 3 release, here.

Posted by David Marshall on April 12, 2006 04:55 PM

April 12, 2006 | Comments: (0)

Welcome Message

Welcome everyone! This is the first post on this blog.

The Virtualization Report is intended to be a place where people can find out the latest news, information, impressions, comments, and reviews about anything relating to virtualization.

The virtualization space is hot right now! New virtualization platforms are being rolled out, software companies are releasing applications to leverage virtualization at record speeds, individual consumers are creating and sharing virtual appliances with one another, the hardware industry is focusing its attention on virtualization within the storage and networking arena, and the IT community in general is embracing the technology more than ever now that it sees the ROI and time savings that it offers.

Folks, I am very passionate about virtualization and everything that goes along with it, and I hope you are as well. This is going to be a wild ride, and I sincerely hope that you not only enjoy the information that I pass along, but that you also actively participate with your feedback and comments. I want this to be as interactive as possible. Your insights and thoughts on the subjects discussed will keep me and the other readers fueled.

Welcome aboard!

Posted by David Marshall on April 12, 2006 03:59 PM

VIRTUALIZATION REPORT PODCAST

Listen to the latest podcast:

MP3 •

•

• Archive •

Technology White Papers

InfoWorld Technology Marketplace

» Technology White Papers Library

Technology White Papers by Topic

Technology White Papers E-mail Alert