One of the most arduous parts of business travel is getting the trips approved, booked and then, finally getting paid back through your expense reports.

Concur is a leading provider of integrated travel and expense management solutions for businesses and they chose HP ProLiant servers based on AMD Opteron™ processors to help consolidate their IT environment.

Concur needed to drive more performance out of less hardware, and obviously virtualization is a great way for businesses to get more out of their IT investments.  Through the use of both VMware VSphere and Microsoft Hyper-V software, Concur was able to reduce their server footprint and get better utilization in their data center. Initially they started with a few servers, and eventually worked their way up to 70% of their environment virtualized.  And according to their statements, Concur expects to get to 80 or 90% virtualized in the future.

Some of the improvements that they saw included reducing the number of servers by 40%, making it easier to manage their environment. In addition, the virtualized environment allows them to increase data by 30% but still maintain the same number of virtualized host systems. Instead of taking four weeks to deploy a new server, they can now do it in 6 hours, helping IT react more quickly to the needs of the business.

All of this resulted in a 10% reduction in their total cost of ownership, a clear business benefit that anyone, especially in these difficult times, can appreciate.

For more details, you can check out the case study on HP.com:

http://h20195.www2.hp.com/V2/GetDocument.aspx?docname=4AA3-4346ENW&cc=us&lc=en

Now, if someone could just figure out how to beam me from Austin to Tokyo, I’d be so appreciative. 

John Fruehe is the Director of Product Marketing for Server, Embedded and FireStream products at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

One of the many things I love about my job here at AMD is that I get to meet and talk with a very diverse group of industry leaders who are tasked with analyzing, planning and implementing the latest and greatest in IT solutions and technologies.  I’ve watched virtualization as both a technology and a strategy evolve and transform how datacenters and even desktops are managed.  I’ve watched the dramatic change in how we evaluate x86 server platforms in terms of performance, power efficiency and overall value and how virtualization has played a key role in this transformation.

The advancement in virtualization technology has now led us to cloud computing.  Recently, I was fortunate to contribute, along with Margaret Lewis, in the writing of the AMD sponsored Cloud Computing Clusters for Dummies book.  In the book we discuss cloud computing in general, what some of the unique server requirements are for cloud computing, and how AMD is driving the adoption of cloud computing as an IT infrastructure.

There is no question now that cloud computing is real – though I wonder sometimes if the real question isn’t “just what is cloud computing”?  It seems to me that the definition of the cloud depends on who you talk to.

Today it is difficult to open a trade journal, read a blog, or even watch TV without finding some reference to “the cloud”.  Mothers are creating, editing and sharing family pictures via “the cloud”.  Business men and women are collaborating on documents, spreadsheets and presentations all through “the cloud”.

We defined “the cloud” in our Cloud Computing Clusters for Dummies book as “the next stage in the evolution of the Internet;  the means through which everything – from computing power to computing infrastructure and applications, from business processes to personal collaboration – can be delivered as a service wherever and whenever needed”.  The cloud is a set of approaches that can help organizations quickly and effectively add and subtract resources in almost real time.  Cloud Computing is about both the business model and the technology.

Some would lead us to believe that in the near future everything will be in the cloud -that the cloud is in fact a replacement for today’s traditional data center.  Prominent industry leaders have predicted that in the future the traditional data center will be replaced by a small number of very large cloud-based datacenters.

Still others are more cautious and are voicing concerns about security, privacy, and regulatory compliance.

Traditionally, Cloud Computing is said to have two distinct models.  The public cloud, like those you hear about from Amazon, Google, Microsoft and others where services are offered on a per use basis and the underlying infrastructure is shared amongst all customers.  And the private cloud, the cloud infrastructure that sits behind the firewall but provides cloud-like services to its select set of internal customers.  More recently we have been hearing talk about a third model – a hybrid model, or the Hybrid Cloud.

According an Information Week interview of HP’s Russ Daniels, CTO for cloud computing and VP of Cloud Strategy, HP’s take is that “virtually every enterprise will operate in hybrid mode, with some of its operations on-premises and some in the cloud”.  Daniels went on to say that contrary to some theories put forth, cloud computing is not a replacement for the data center.

This seems to be a much more practical approach to me.  Yes, some small businesses may find it practical and cost-effective to move all of their IT processing to the cloud, and certainly all of us will use cloud-based services (think social media and email) at an increasing rate.

But the enterprise is a much more complex environment.

I found this definition from TechTarget’s website SearchCloudComputing.com:

A hybrid cloud is a cloud computing environment in which an organization provides and manages some resources in-house and has others provided externally. For example, an organization might use a public cloud service, such as Amazon Simple Storage Service (Amazon S3) for archived data but continue to maintain in-house storage for operational customer data. Ideally, the hybrid approach allows a business to take advantage of the scalability and cost-effectiveness that a public cloud computing environment offers without exposing mission-critical applications and data to third-party vulnerabilities.

Both Microsoft CEO Steve Ballmer and Paul Maritz, president and CEO of VMware appear to agree on this concept.  In a recent interview with ITworld, Mr. Ballmer was quoted as saying “I think the new solutions that get pioneered in the cloud will also get retrofitted so they can be run on-premises by customers, because no customer is going to have everything in the cloud. People are going to have these kinds of mixed environments”.

Similarly, Mr. Maritz recently noted, “The new service based on VMware vCloud Datacenter enables enterprises to not only improve efficiency internally, but also be in a position to tap into external resources in a non-disruptive way when it makes business sense to do so.”

The cloud is indeed real, and my guess is that in the future most if not all enterprises will realize the benefits of some form of hybrid model.

I’m interested in hearing your thoughts on cloud computing and the hybrid cloud model specifically.  Are you moving any of your enterprise applications to the cloud?

Tim Mueting is a Product Marketing Manager at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

Where is Tinkerbell when we need her?  Software licensing definitely feels like a “Never Never Land” topic.  At the heart of the issue is the need to mature the business models around the multi-core, scale out clusters that are central to virtualization and cloud computing.

At the conference, Gartner analysts noted that virtualized software licensing continues to present a major stumbling block to widespread virtualization adoption. They believe organizations that do not diligently monitor how software vendors are responding to virtual use issues are likely to experience significant cost increases and the unintended impairment of their current license rights.  At AMD we continue to battle the misguided perception that more cores automatically means higher licensing costs. To that end, we encourage end users to talk directly with software vendors about multi-core licensing methodologies.

As a refresher, there are several ways in which server software is licensed, including by:

• Processor socket
• Number of processor cores
• Number of VMs on a server
• Physical Server (regardless of number of processor sockets, cores, or VMs)
• Users accessing the server (either number of users or by named users)
• Service contract (yearly or multi-year)
• Site (all servers included in a location)
• Volume license agreement (sometimes call Unlimited License Agreement)

Note that the majority of server ISVs, including Microsoft^® (SQL Server, Exchange, Windows Server^®, Hyper-V), Red Hat, Citrix, and Parallels, do not license by core count.  The server ISVs that offer some type of software licensing scheme based on core include IBM, Oracle, VMware and certain HPC software.

It’s also important to remember that it’s rare for a software vendor to offer only one type of licensing. For example, Oracle has published licensing schemes that are based on core count.  In that instance, AMD seems to be disadvantaged as the price of the software is much higher on AMD-based servers because of our higher core count.  However, if you visit the Oracle Wiki on software licensing, you find a complete list of the licensing schemes offered by Oracle. And most customers tend to utilize the site or enterprise licenses when using Oracle software, rendering core counts immaterial.

VMware is another ISV that is perceived to have unfavorable licensing methods for larger core counts.  Given the many articles written about VMware licensing, it’s no surprise that the topic remains top of mind in the industry. VMware’s vSphere Standard and Enterprise editions only support processors with up to 6 cores. This means AMD and Intel processors with larger core counts require the purchase of Advanced or Enterprise Plus editions.  Keep in mind that AMD-based servers offer excellent price points that often nullify the higher cost of the Enterprise Plus license.  Further, refer to my “Simply Spectacular Virtualization” blog of July 6, 2010 for a sample case study of server and software pricing. And, let’s not forget that in addition to the price being nullified, the Advanced and Enterprise Plus versions offer more advanced enterprise features, so while the net price (HW+SW) turns out to be about the same, the AMD solution actually has more features.

Another way to skirt the issue for AMD Opteron™ 6100 Series users is to buy the eight-core version and turn off two cores per processors.  You might be asking yourself “why buy cores only to turn them off?”  Well, in many virtualized environments memory bandwidth becomes more of a limitation than CPU performance.  AMD Opteron 6000 Series platforms offer increased memory bandwidth over competitive systems, as seen in the results from STREAM memory bandwidth benchmark. It should also be noted that Intel limits the QPI I/O speed on many of its low-power Xeon processors.  For example Intel’s L5640 is limited to 5.86GT/s – while the AMD Opteron 6100 Series processor can handle up to 6.4 GT/s on all of our processors.

Starting on September 10, 2010, VMware began migrating licensing of several of its products to a “per VM” model. While VMware vSphere continues with its existing licensing model, it is anticipated that VMware will need to update vSphere licensing to better reflect larger core count processors.

AMD continues its efforts to develop “real cores” and address the needs of scale out clusters.  A blog by my colleague John Fruehe helps to drive home the importance of having more cores.  As we continue our march toward 2011 and the release of processors based on our new “Bulldozer” core, we will continue to work with software vendors to explore alternate licensing methods that don’t penalize users as they explore the merits of more cores.

The bottom line?  While some point to software licensing as an issue for AMD Opteron processors, the vast majority of the enterprise licenses tend to be platform neutral.

What are your thoughts on software licensing? Do you feel the industry is moving fast enough to resolve the licensing issues raised by multi-core and virtualization?

Margaret Lewis is the Product Marketing Director at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only.  Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

Not only did we solicit your participation in our online contest, we also met a number of you at VMworld in San Francisco.   A few of you were passionate enough about the subject to jump on camera to share your opinions and predictions. With VMworld Copenhagen this week, we thought it was a great time to share your video responses.

www.youtube.com/watch?v=9Rosv1KzT0k

Much like our online respondents, VMworld attendees defined the cloud as an end to the constraints of computing as we’ve come to know it.  They were excited about the possibilities of combining platforms, operating systems and applications that can benefit from the cloud computing model.  As we expected, respondents are using the cloud both in a personal and professional capacity – in more ways than we could even begin to count.

www.youtube.com/watch?v=O7KmtAuWeFg

Then we asked– how do you define the ideal cloud infrastructure? Over and over again we heard: the ideal cloud infrastructure is scalable and reliable.  We agree, and our recently launched AMD Opteron 4000 Series Platform is designed to address exactly those requirements.

So, how did our respondents do, and is there anything they missed?  If you’re at VMWorld in Copenhagen, we’d also love to hear from you.  Is the cloud computing hype in full force on another continent this week?

Tracey Blanton is a Product Marketing Manager at AMD.  Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

But, when delivering a  product to market, what people tend to focus on – naturally – is the end result. The blood, sweat and tears that went into making the product a reality can sometimes get overlooked. But, not at AMD.

From start to finish, we’ve been working in lock step with VMware and their engineering group to ensure our new AMD Opteron^TM platforms would be superb. I can say without hesitation that this partnership was instrumental to our success, and in fact, we couldn’t have done it without them.

With that said, on Friday, August 27, AMD threw a well-deserved “Thank You” party for 350 VMware engineers who supported the delivery of these products.  The festivities included a custom cake (VMware and AMD do make cloud computing delicious), Italian gelato, and test flights of indoor RC Helicopters (Next time maybe we’ll invite a few air traffic controllers to join the fun!). 

Thanks VMware! 

John Fritz is a Director of Strategic Software Alliances at AMD.  His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

So what does the trend toward virtual desktops mean to me – your garden variety road warrior/knowledge worker? The promise of fewer maintenance hassles and a more reliable way to keep my data safe and accessible. In this regards both IT and I have the same goals – something that does not always happen. However, the world of virtual desktops has not always offered the most satisfying user experience (my fear) and the ROI is often not as straight forward as server consolidation (IT’s fear).

The virtual desktop moved closer to reality for both users and IT with the launch today at VMworld of VMware View 4.5. For me, the user, VMware View offers full desktop capabilities, even while disconnected from the corporate network. View can also deliver a personalized desktop experience across a variety of end-user devices, including Windows and Mac devices, as well as thin clients – a key feature since the number of computing devices I have seems to be growing instead of decreasing.

From an IT standpoint VMware View 4.5 brings a managed desktop virtualization package that can support a massive number of desktops through a single console. VMware View 4.5 is designed for scalability, flexibility, and performance – all attributes shared by the AMD Opteron™ platform.  In fact, the latest VMware desktop solution is designed specifically to take advantage of multi-core processors, such as those offered by AMD, to deliver an excellent user experience. 

VMware View 4.5 can be a great fit for customers deploying desktop virtualization with the new AMD Opteron ™ 6000 Series platform which is designed to increase VM density and support more virtual desktops per server. Based on the next-generation AMD Direct Connect Architecture 2.0, the new AMD Opteron 6100 Series processors offer:

• Up to 24 cores in a 2P configuration or up to 48 cores in a 4P configuration
• Up to twice the memory bandwidth as the previous generation of Operton processors, with 12 DIMM slots per CPU
• Improved power efficiency features with hypervisor support for Enhanced AMD PowerNow!™ and C1e state
• AMD-Vi™ technology for I/O Virtualization with the AMD SR56x0 chipset
• Outstanding price-performance balance as compared to competitive offerings

As the first 8- and 12-core x86 processor for the 2P and 4P server markets, the AMD Opteron 6000 series platform offers double the core count of AMD’s previous generation 4- and 6-core processors, which helps deliver greater VM densities and more virtual desktops per physical server.  Add on top of that a competitive price, more memory channels than the previous generation and improved power features, and you have an environment perfectly positioned for the next generation of desktop virtualization, aided by the launch of VMware View 4.5.  

We have a long history of collaboration with VMware, extending back to before the original AMD Opteron processor launch in 2003.  At that time, server virtualization on x86 computers was in its earliest stages and server-hosted desktop technology, referred to under the mainframe-style name of “terminal serving”  was making its way into the x86 computing world. Today, thanks to our partner VMware, server virtualization is moving into the mainstream, as nearly every enterprise IT shop has implemented some level of virtualization, and desktop virtualization is gaining more and more attention as the technology matures.

Check out more information on VMware View 4.5 here.  

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

A lot has been written recently on the number of virtual machines (VMs) that can (or should) optimally and safely run on a single physical server.  While there are many differing opinions, most agree on the basic principles.   The customers I talk to tell me they attempt to balance performance, utilization and risk in determining what is “optimal”.

We all want to get the most out of our hardware investment, and server consolidation through virtualization plays an important role in meeting, and in many ways exceeding, this goal. But determining just how many VMs can be supported on a particular hardware platform without impacting the performance needs of end user applications can be a tricky task.  We have to take into consideration the needs of the differing workloads sharing the same physical resources, including CPU, Memory or I/O bandwidth.

Virtualization can place a high demand on the server’s CPU, memory, cache and I/O bandwidth.   To improve just how those resources are used and shared in a virtual environment – in some cases taking the burden off managing those resources in software – AMD has implemented several CPU extensions with our AMD Virtualization™ technology.  In addition, the added number of cores, now up to 12 cores per socket with our AMD Opteron™ 6100 Series processor, the additional memory capacity and memory channels, and faster Hyper Transport Links (think HT3) have all helped to increase VM density, or the number of virtual machines that can be on a single physical server.

Based on a very unscientific, random poll I’ve been taking at various trade shows, user group meetings and customer visits, it appears that most administrators attempt to stay within 50 – 65% capacity of their physical servers as long as they are meeting their performance requirements.  This seems very realistic to me and a far cry better than the 3-8% utilization rates we were getting prior to implementing a virtualization solution.

At the same time, they tell me they are still reluctant to overload a physical server with a large number of VMs even if the capacity is available. What happens if a hardware failure is encountered?  More VMs would be unavailable, more applications would be unavailable, more users would be affected and it would take longer to move those VMs to an available physical server and get everything up and running again.  Therein lays the virtual balancing act.  We still have to plan for a suitable time to recover in case of an unplanned outage.

Cores and Memory

Another important aspect when considering how many VMs to run is memory.  How much memory do you need per core or per VM?  It’s my understanding that some experts recommend 2-4GB of memory per core, while others recommend that amount per VM depending on the type of workload. We know from experience that the more cores and memory we have for our virtual environments, the more capacity and better performance we can achieve.

And we can’t forget about I/O.  There are new features and technologies being introduced to ensure better throughput and performance for those workloads with heavy I/O characteristics.  Stay tuned here for more on that topic soon.

Resources

There are several tools and technical resources available that can help give you a better idea of what your configuration should look like.  VMware offers a product called VMware vCenter CapacityIQ which plugs into vCenter Server and provides capacity management for both datacenter and desktop environments.  Microsoft offers its Assessment and Planning Toolkit which provides server virtualization scenarios to assist in planning for server consolidation with Microsoft Hyper-V.

HP, Dell and a variety of others offer services that utilize these same tools and various other resources that can also be effective, in addition to providing you an independent evaluation of your needs.

Back to our original question: how many is too many?  According to a 2008 study by SearchVirtualization.com, 62% of the respondents were running 10 or fewer virtual machines per host and 5% were running more than 25 VMs per host.  I’m sure these numbers have increased over the last year or more, but by how much?

What criteria do you use to optimize your environment?  How many VMs are you running today and how many do you consider ideal?

Tim Mueting is a Product Marketing Manager at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

Virtualization has only added to this.  Consider just one part of the issue, what constitutes a relevant virtualization benchmark?  In our opinion, a relevant virtualization benchmark should include a combination of different workloads with differing characteristics running on an increasing number of virtual machines.  This approach will sufficiently stress the hardware and the software stack to determine what we think will be the optimal configuration for performance and efficiency.

Up until now, VMware’s VMmark benchmark has been the industry’s adopted standard for measuring performance of the VMware hypervisor.  I’ve shared with you before my assessment of VMmark and its limited application to real-world environments, and I’ve shown where different methodologies can and will produce vastly different results.  In the end, when analyzing performance, the answer is often the dreaded “it depends”.

We now have another viable option to assess and plan for our virtualization environment.  On Wednesday, July 14^th, SPEC released the first hypervisor-neutral virtualization benchmark for servers, which measures performance, power consumption and the relationships between power and performance.   According to SPEC, SPECvirt_sc2010 uses a realistic workload and SPEC’s performance and power measurement methodologies to enable vendors, users and researchers to compare systems performance across multiple hardware configs, virtualization platforms, and applications. The fact that SPECvirt is “vendor neutral” in terms of the hypervisor is a real plus for customers who are trying to evaluate all aspects of the virtualization platform. It also shows the maturation of the virtualization market to include competitive offerings from major software vendors like Citrix, Microsoft, Red Hat, and VMware.

SPECvirt_sc2010 was developed by the SPEC virtualization subcommittee of which AMD was a key contributor, along with many other of our partners like Dell, HP and VMware.  This new benchmark modified several existing SPEC workloads; SPECweb2005, SPECjAppServer2004, and SPECmail2008, all common targets of virtualization, to match a typical server consolidation scenario such as CPU resource requirements, memory, disk I/O and network utilization. 

AMD is proud to have been a part of the development and release SPECvirt_sc2010 and we look forward to seeing it evolve in the industry.  We believe the addition of a power metric and the ability to analyze power/performance relationships at both the server and the system level is big step forward in the area of virtualization benchmarking.

SPECvirt_sc2010 doesn’t come free.  A license can be obtained from SPEC for $3000. Membership in SPEC is open to any interested company or entity that is willing to commit to SPEC’s standards.  More information is available at http://www.spec.org/virt_sc2010/.   

So where does this leave VMmark?  Only time will tell.  Have you looked at SPECvirt_2010?  I’d love to hear your thoughts.

Tim Mueting is a Product Marketing Manager at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

The way I see it, as the software continues to get smarter and smarter, and the hardware evolves to be more and more efficient (take our latest AMD Opteron™ 4000 and 6000 Series platforms), cloud computing is becoming much more accessible to customers. This is a huge step for the entire IT ecosystem.

Yet while cloud computing itself may be trending towards greater accessibility, management of the cloud is becoming more and more complex.  To help address this issue, VMware has broadened its vCenter management portfolio with VMware vCenter Configuration Manager and VMware vCenter Application Discovery Manager.  These products are now more tightly integrated with VMware vSphere providing a more efficient management solution.  Customers now have more tools to help to improve the efficiency of managing their virtualization environments through an automated, policy-based solution.

In addition, VMware vSphere 4.1 also includes new performance and scalability enhancements, including a new memory management technique called “memory compression” and expanded resource pooling capabilities that aggregate pools of resources which, according to VMware, provides up to twice the scalability as the previous version.

The release of VMware vSphere 4.1 couldn’t come at a better time. The demands for a cloud-based delivery model are greater than ever as data explodes around the world, and information abounds online (some stats on that data explosion in this video). In order to effectively handle all of that data without exponentially increasing power consumption or data center real estate costs, innovative solutions – such as public and private clouds – become a necessity. AMD has worked with VMware for years to ensure the hardware and software deployed in the datacenter work together seamlessly, ensuring that innovation doesn’t happen in a silo

As datacenter technology becomes more and more sophisticated, and the cloud continues to mature, what are the pain points you’re still facing? We welcome your comments here, and hope you add to the discussion.

Tim Mueting is a Product Marketing Manager at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

 The Dell PowerEdge R815 is a great example of a value 4P server.  Dell has packed a lot of capability in a 2U form factor. Designed to support up to 48 processor cores, the Dell PowerEdge R815 offers a balanced architecture of advanced manageability, memory scalability of up to 32 DIMMs, massive I/O, and system redundancy – all in a space-saving 2U form factor. You get an option of AMD Opteron processors with either 8 cores or 12 cores along with three different power bands – all with the consistent feature set.

What does it cost to own this value 4P server? Let’s check out the cost compared to a competitive 2P system.  Pricing was derived from the hardware vendors’ on-line system configuration tool for SMB servers as of June 3 and 4.  Systems take into consideration the OEM’s recommendations for memory and power configurations and include one 160GB SATA hard drive and a basic level of service.

The chart provides the hardware cost (system cost) and hardware + virtualization software cost (system +VMware vSphere). There are a couple of points to note. The cost of the competitive 2P server is lower – but so are its capabilities. I am comparing a 2P server with 12 cores, 64GB of memory, and VMware vSphere Enterprise version  to the value 4P server with up to 4x the number of cores, double the amount of memory, and the Enterprise Plus version of VMware vSphere 4. It should be noted that the cost of the 2P competitive server would be about $10,000 more if it was configured  with 128GB of memory and the vSphere Enterprise Plus edition like the Dell PowerEdge R815.

So let’s go beyond the raw cost figures and look at these servers with Virtual Machine (VM) deployments. There are two evaluations of “cost per VM.” The first one assumes that each VM will have 1 vcpu and the total number of VMs for each system matches the total number of physical cores. This logic matches the methodology used by some of our major web hosting customers and is also a more traditional configuration for hosted desktops sessions with software like VMware View or Citirx XenApp.

The second evaluate assumes that the VMs each have 4 vcpus and the total number of VMs for each system is the total number of cores/4. This type of configuration might be used for a more robust workload like a Microsoft’s SQL Server or a business application like SAP.

In either case – the end result is the same. The Dell PowerEdge R815 blows past the competitive 2P server in terms of capabilities and cost per VM. With a savings of over $6,000 ($49,272 for two HP ProLiant DL 380 G7 servers as compared to $42,574 for one Dell PowerEdge R815 server)this shows the philosophy of the value 4P server  – a simply spectacular platform for  virtualization!

My colleague, John Fruehe, published a blog The Value 4P – Courtesy of the AMD Opteron™ 6000 Series Platform, that provides an overview of this new type of server.  And I just recently participated with Dell and VMware in an Information Week webcast Performance & Value: Virtualizing with the Dell PowerEdge R815 Server  that provides some more insights into the value 4P server.

 So let me know your thoughts. Would you consider a value 4P server as a virtualization platform for your data center?

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

Dylan had it right in 1963 when he recorded this classic that some say captured the spirit of social and political unrest of the times.  Today, the times certainly are a-changin’, especially when it comes to the computing power of x86 servers.  The role of x86 servers in the world of enterprise IT continues to grow.  In fact, according to an IDC report, the first quarter of 2010 experienced the fastest year-over-year revenue growth for x86 servers in more than 10 years.  Virtualization, once a niche technology is growing more pervasive in today’s datacenters. 

Servers based on x86 technology are equipped with unprecedented amounts of I/O and memory capacity.  The latest AMD Opteron™ 6000 series platform (formerly known by the code name “Magny-Cours”) is based on the AMD Direct Connect Architecture 2.0.  The platform now boasts up to 12 cores per processor, supports up to two times the memory and I/O bandwidth of the previous generation Six-Core AMD Opteron™ processor and includes AMD-V™ I/O Virtualization technology in the chipset.  

With all the advancements in virtualization technology over the past few years, it appears we’re ready to tackle what some are calling the third leg of server virtualization – the network.  A great example of this is HP’s latest release of Virtual Connect Flex-10 networking adapters for the HP ProLiant BladeSystem, and VMware vSphere vNetwork Distributed Switch feature.  HP’s Virtual Connect Flex-10 is hardware-based I/O technology that enables you to partition a 10Gb Ethernet connection into four virtual connections and fine-tune the bandwidth of each of those connections.

Virtual Network Design

We’ve seen how virtualization has changed the way servers are designed, and we are now seeing virtualization change the way networks are designed.  In the world of virtualization, a physical server hosts many virtual servers, which can now communicate with multiple virtual network interface controllers (NICs).   Virtual switches contained within the physical server extend the network infrastructure beyond the physical NIC which creates a virtual network infrastructure – all of which must be managed and tuned.

Enter HP Virtual Connect Flex-10.  Virtual Connect Flex-10 defines a physical end-point to the network infrastructure and allows a server administrator to control the assignment of virtual networks (or VLANS) to the physical NICs inside the BladeSystem.  This Virtual Connect end-point clearly separates the role of the server administrator from the role of the network administrator. The network administrator provisions the network VLANs to the Virtual connect infrastructure and the server administrator connects the NICs to the appropriate network.

Virtual Connect works with VMware’s  vSphere 4.0 vNetwork Distributed Switch (vDS).  vDS extends VMware’s legacy vSwitch technology with features that enable spanning multiple hosts with a single virtual switch.  vDS now treats the network as an aggregated resource. Individual, host-level virtual switches are abstracted into a single large vNetwork Distributed Switch that spans multiple hosts at the datacenter level.  While a vNetwork Distributed Switch can plug into any existing virtual infrastructure, HP Virtual Connect Flex-10 provides a mechanism for presenting consistent “networks” and VLANs to all hosts within the ESX cluster.

With vDS, VMware has also added a number of other new capabilities, not available with Standard Switches, that address emerging virtual network requirements.  A few of the more interesting features are:

• Private VLAN (PVLAN) enables users to restrict communication between virtual machines on the same VLAN or network segment which helps to address security concerns and help reduce the number of subnets needed for certain network configurations.
• Network VMotion tracks the virtual machine networking state as the VM moves from host to host on a vNetwork Distributed Switch providing a consistent view of a virtual network interface regardless of the VM location.
• Bi-directional Traffic Shaping helps users limit the traffic to or from a VM or group of VMs to protect a VM or other traffic in an oversubscribed network.

Software Licensing

With the introduction of vShpere 4.0, VMware introduced a new tiered licensing model.  Those of you looking at taking advantage of VMware’s vNetwork Distributed Switch will need to upgrade to vSphere Enterprise Plus. 

You can now enter into a new generation of Virtualization with technology like HP Virtual Connect Flex-10, vSphere vNetwork Distributed Switch and HP ProLiant BladeSystem servers equipped with AMD Opteron™ 6000 series processors – designed to deliver unprecedented price/performance, vastly improved manageability and decreased power and cooling costs.

 Are you considering network virtualization technology for your data center?

Also see John Fruehe’s blog “The Value 4P – Courtesy of the AMD Opteron™ 6000 Series Platform for an overview of the Magny-Cours platform, and check out my blog Understanding the Bigger Picture of VMmark Benchmarks for a review on the latest virtualization performance results.

Tim Mueting is a Product Marketing Manager at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

Many providers are trying to join the virtualization discussion by simply focusing on the components of the technology. However, VMware is taking a different approach by keeping the focus on you, the customer. AMD, along with its hardware partner Dell, is proud to sponsor VMware Express and showcase PowerEdge servers featuring AMD Opteron™ processor technology.

Here’s a list of upcoming locations where you can see AMD and Dell live and in-person.

• May 26: Scottsdale
• June 4: Houston
• June 16: Nashville
• June 22: Dallas / Ft Worth

Be sure to check the website for additional information as the tour progresses: www.vmware.com/tour

The VMware Express is a great way to get educated about the many performance, efficiency and security benefits of virtualization and cloud computing. By boarding the VMware Express, you can gain:

• Hands-on experience with VMware’s award-winning desktop virtualization and cloud computing solutions in real-world scenarios.
• Insight from technical experts about the benefits that virtualization and cloud computing solutions can provide to specific businesses, from the desktop through the datacenter, and to the cloud.
• Access to five demo stations that illustrate VMware desktop and server solutions, including VMware View™, VMware vSphere™ and the VMware vCenter™ family of products.

You might also want to take a look at a success story that showcases the City of Austin and their use of Dell PowerEdge Servers running AMD Opteron processors with VMware ESX to help prune power and cooling requirements while boosting testing and development capabilities.

Virtualization and cloud computing have relevance for businesses of all sizes.  Before making an investment in these technologies, however, take a few minutes to kick the tires and test drive available solutions. We hope to see you soon at a location near you.

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

First, it is important to understand that there are a wide variety of licensing methods used by server software and most ISVs offer several options to their users. These methods include:

• Number of physical processors or processor sockets
• Number of cores
• Named users  or client access license (CAL)
• Volume or site license
• Deployment (can be several computers)
• Annual subscription fee

Next we need to take a quick look the licensing methods used by popular server software packages:

• Microsoft server software products, including Windows Server 2008 R2/Hyper-V R2, SQL Server 2008, and Exchange 2007, use a variety of licensing methods including by processor socket, by CAL, or a combination of these two.  You can also negotiate volume licensing directly with Microsoft. It should be noted that Microsoft software products do not license by core.
• Red Hat Enterprise Linux uses an annual subscription agreement that is based on the number sockets in a server.
• Oracle database licenses by processor, core count, user, or a negotiated volume license agreement. Oracle applications typically license by user or some type of volume use methods.
• VMware licenses vSphere by processor or by processor/core count based on the edition you buy, and each edition offers a different set of features/options.

So let’s drill down on two software packages that generate many “license by core” questions -  VMware vSphere and Oracle database.

With the introduction of VMware vSphere 4, VMware made a change to its licensing method. There are now several editions of vSphere, each offering a defined set of features/options that were previously purchased “a-la-carte.”

• The Standard and Enterprise editions license by processor but limit the number of cores that each processor can have to 6 or below.
• The Advanced and Enterprise Plus editions are licensed by processor and don’t put any limitations on core count.
• Since each version has a specific feature set customers unsure about which edition to purchase can use the vSphere Purchase Advisor to get a recommendation.
• Customers who want to explore other licensing methods with VMware should contact them directly.

There are three editions of Oracle database software: Standard Edition One, Standard Edition and Enterprise Edition. Here is what customer’s need to keep in mind with these editions:

• For Standard Edition One and Standard Edition, a processor is counted equivalent to an occupied socket; however, in the case of multi-chip modules, each chip in the multi-chip module is counted as one occupied socket. This means that for the AMD Opteron 6100 Series processor, which is considered to be an MCM processor by Oracle, Standard Edition One will only support 1 socket servers and Standard Edition will only support 1 or 2 socket servers.
• For Enterprise Edition, the number of required licenses is determined by multiplying the total number of cores of the processor by a core processor licensing factor specified on the Oracle Processor Core Factor Table. For AMD Opteron processors this factor is currently .50, so ($47,500 x # of cores on server x .50) is the equation used to calculate costs.
• Customers who want more details on Oracle database licensing or want to explore other licensing methods such as user of volume licensing should contact Oracle directly.

Finally, it is also helpful to look at software licensing from a broader data center perceptive and not just on the individual server level. For example, in the article “Software Licensing Costs May be Reduced by AMD’s 12-Core Chip Released Today,” Matt Lavallee, the Director of technology at MLS Property Information Network Inc., takes the view that upgrading from his current quad-core chips to the 12-core AMD Opteron 6100 Series processor can let him cut down the number of physical servers, which in return can cut down his overall software licensing costs.

AMD continues to work closely with its ISV partners to continually evaluate software licensing issues. We always stress that customers need to work directly with the ISV so they are aware of your unique concerns and needs.

We look forward to hearing your thoughts on the world of software licensing and the high anticipation of our AMD Opteron 6000 Series platform. Things are certainly getting exciting!

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

Microsoft:

“Microsoft has partnered with AMD in the data center since the original AMD Opteron processors were launched in 2003, and we naturally have been working together to deliver advanced efficiencies with the new 8- and 12-Core AMD Opteron 6000 Series platforms,” said Eric Jewett, Director, Microsoft Server Marketing.  “Customers running Windows Server® 2008 R2, Hyper-V R2, and SQL Server can benefit from the balanced price, performance, and power these new server platforms offer.”

VMware:

“With AMD Opteron 6000 Series platforms, AMD is taking a big step forward in terms of efficiency and performance-per-watt. Our flagship virtualization solution – VMware vSphere™ 4 – is fully compatible with and optimized for ‘Magny-Cours’,” said Richard Brunner, chief platform architect and principal engineer, VMware. “Combining VMware’s proven virtualization technology with AMD’s most impressive server platform to date helps our joint customers to fully harness the promise of virtualization.”

Red Hat:

“For years, AMD and Red Hat have been focused on providing outstanding reliability and flexibility in the datacenter,” said Tim Burke, vice president, Engineering, Platform Engineering at Red Hat. “AMD’s next-generation 8- and 12-Core AMD Opteron 6000 Series platforms extend that focus, and when paired with Red Hat Enterprise Linux®, these server platforms can help enterprise customers achieve impressive performance, while also helping to maximizing their IT investment.”

Parallels:

“Parallels and AMD began collaborating to provide virtualized solutions to the web hosting community well before cloud computing was a buzz word, and we are excited to continue the tradition of mutual innovation with the release of the new AMD Opteron 6000 Series platforms,” said Amir Sharif, vice president of Virtualization at Parallels.  “Both Parallels Virtuozzo Containers and the new Parallels Server 4 Bare Metal are fully compatible with AMD’s latest server platforms, and we look forward to bringing improved, highly efficient virtualization solutions to customers around the world.”

Citrix:

“Together, Citrix and AMD have been helping to pave the way in the industry for efficient and optimized client and server virtualization solutions to meet the demands of today’s IT environment,” said N. Louis Shipley,  group vice president and general manager, XenServer Product Group, Citrix Systems. “Now, with the availability of platforms based on the latest 8- and 12-Core AMD Opteron 6100 Series processors, we’re able to deliver further enhanced scalability and performance for virtualization through Citrix XenServer and XenApp.”

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

But all this talk about where the cloud is going got me thinking about where we’ve come from. In fact, it was almost exactly a year ago that I was talking about getting beyond the fluff of cloud computing, and eWEEK was asking if we can “bet on the cloud.” The evolution of cloud computing allows you to draw a number of parallels to similar technologies. Living and breathing virtualization, I can’t help but feel some déjà-vu when I hear the main issues of cloud computing discussed (both pros and cons): it’s incredible for efficiency, flexibility and scalability of your business, but what about manageability, security and overall maturity of the market? These are all things we’re saying about virtualization just a few short years ago.

But, like virtualization, the major players stepped up to drive cloud computing forward. VMware launched their vSphere operating system – a major step to addressing those concerns over manageability & security. Microsoft has launched their Azure platform, which helps simplify the entire development cycle of the cloud – from web developers to ISVs to system integrators. And, from a standards perspective, groups like the Cloud Security Alliance and the Distributed Management Task Force (DMTF) have taken control of this market to help ensure its evolving in a responsible way.

And, at AMD, we’ve been addressing the infrastructure needs. We now have a Six-Core AMD Opteron EE processor available that draws an ACP of 40W, as well as the “Kroner” server platform to further address energy efficiency. These low-power, highly scalable, fully-featured solutions help cloud providers operate an extremely efficient, highly utilized and dense computing environment.

The same way that AMD was ahead of the curve in developing on-chip virtualization features like Rapid Virtualization Indexing (RVI) and focusing on low-power server processors to ensure a highly utilized physical machine was saving energy costs – AMD is out in front with cloud computing again. The conversation has evolved from “will this catch on” to “how can we shift our business model to adapt,” and it’s conferences like the two happening this week that are instrumental in driving that discussion.

What do you think? Are we finally beyond the fluff?

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD.

Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

Low and behold – I was not the only person working on this thought thread. We recently sent Johan De Gelas with AnandTech a 4P Six-Core AMD Opteron processor-based system.  And in his review, “Expensive Quad Sockets vs. Ubiquitous Dual Sockets,” Johan focuses on some interesting points on the same topic (which my colleague John Fruehe discusses on the @Work blog as well). While he compares a number of workloads on 2P servers versus 4P servers- my interest was drawn to the section where Johan discusses one of my most favorite topics – virtualization (surprise, surprise).

In my research I have been taking a look at the pricing of 2P and 4P servers with larger memory configurations. Here is a snapshot of system costs using processors with similar specifications. Note how steeply the system price climbs for the 2P servers as the memory configurations grow.

Johan came to a similar conclusion  ”A very large database or virtualization consolidation scenario requiring more than 72GB of RAM will probably push you towards the quad Istanbul – once you need more than 64-72GB, memory gets really expensive on the Intel dual socket platform. There are two reasons for this: 8GB DIMMs are five times more expensive than 4GB DIMMs, and DDR3 is still more costly than DDR2 (especially in large DIMMs).”

OK – this is a compelling cost message. But what about performance? Johan gives two examples of performance – current VMmark scores with systems that have memory configurations ranging from 64GB to 128GB, and vApus Mark I, their own in-house developed virtualization benchmark, designed to measure the performance of “heavy” performance-critical applications and done in cooperation with the Sizing Servers Lab.

For performance details, latest results, and system configurations see http://www.vmware.com/products/vmmark/results.html.

So compelling performance - and more importantly – very compelling price/performance.

I did one other exercise during my research. Instead of just comparing server to server – I took a look at comparing a 42U rack of 2P servers versus a 42U rack of 4P servers. Using the system costs from the example above here is another snap shot

The price difference of servers to fill a 42U rack is less than $10,000 when comparing 2P and 4P servers with 64GB of RAM – and over $400,000 more for 2P with 128GB of RAM compared to similarly configured 4P servers.

So the take away from my research – 4P servers running Six Core AMD Opteron processors provide the cores, memory and price to achieve simply spectacular virtualization! And, as I noted, to get some additional insights on the same topic, check out my colleague John Fruehe’s recent blog “AMD Opteron Processor – Delivering True Value, Not Just Benchmarks.”

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD.

 Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

That is 25% more virtual machines than the competition can claim running the same workload. As a side note, running the same workload on either a Six-Core AMD Opteron processor-based HP ProLiant DL585 G6 server or BL685c G6 blade -- we’re able to consolidate 120 virtual machines, which is 43% more VMs than the competition on a 4-socket, 24 core server.

But, don’t take my word for it, check out the video below (we also just broke our own record with HP releasing a a new top 48-core VMmark result on the same server with a score of 53.73 running 210 VMs!).

And, join us at VMworld August 31-September 3^rd. We’ll be showcasing this demo at our booth (#1408) as well as many more.

www.youtube.com/watch?v=mLioRETRpJ4

Tim Mueting is a Product Marketing Manager at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

¹Based on 30 tiles x 6 VMs for 48-core HP ProLiant DL785 G6 server, as tested using the VMmark benchmark (http://www.vmware.com/products/vmmark/results.html).

While there is a lot of talk out there about the benefits of virtualization and cloud-models, a lot of business are asking where the heck you are supposed to start. Let’s slow down and understand something first: one size does not fit all.

Something I’ve become a big proponent of is what I’ll call a hybrid model. What I mean is that you can think about virtualization and cloud computing in different pieces, and integrate those solutions without overhauling your entire infrastructure. In the spirit of practicing what you preach, we saw this success within AMD’s own Sunnyvale campus. In our IT department in Sunnyvale, about 100 engineers deployed server-hosted virtual desktop infrastructure (VDI) so that they could share resources and a common desktop management infrastructure. The desktop virtualization runs as part of AMD’s HP blade server farm and storage environment consisting of HP’s EVA 8100 SAN and HP BL685 and BL465 blade systems with AMD Opteron^TM processors.

This method of desktop virtualization provides users with access to server-based applications and related data via sessions running from central servers in the data center. We are currently able to host about 50 virtual desktops on one server! This has been a simple measure to improve the efficiency of our IT department, who could then provision the desktops of employees to more effectively manage issues without actually having to be at the physical computer. This VDI virtualization implementation increased responsiveness to ongoing project demands while helping to save operations costs.

Utilizing VDI within AMD is just one example. You can find a myriad of benefit from adaptive server and desktop virtualization. For example, security and management. In the VDI example data and applications reside on the server, not the client, which contributes to a more secure environment and on that is more easily managed by IT. It also enables more efficient software provisioning and patching. And the good news – you don’t have to implement expensive client systems with a lot of proprietary technology like Intel’s vPRO to reap these benefits.

During VMworld Conference, we plan to feature presentations from Citrix, Microsoft, and Wyse in the AMD “Masters of Virtualization” Theater, providing insights into this emerging world of desktop virtualization.  Our own Tim Mueting will also feature a customer case study on desktop virtualization during his VMworld session – check out his blog for detail.

So what can you take from all this? Technology is made to empower you, not confuse you. Virtualization and cloud computing are receiving a great deal of attention because of the potential they offer to improve efficiencies and maximize resources, but there is always room for a hybrid approach. In fact, AMD’s very own Neil Kelly was at BriForum last month talking up this exact topic with Wyse Technology. In particular, they discussed solutions that are aligned at simplifying access to mixed data center cloud environments with superior user experience quality.

In this economic climate, I always encourage playing it smart. But, don’t just take my word for it – leave your experiences in the comment section.

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD.

Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

First, let’s take a closer look at how IT managers are actually using four-socket servers. It’s very rare for four-socket servers to be used strictly as a way to increase compute density. Appro and SGI have twin servers and half-depth servers to serve this purpose. In most cases, four-socket servers are used for applications that require a single server to have access to large amounts of memory. These applications typically consist of database, virtualization/consolidation, and some high performance computing applications. With the ability to support 32 memory DIMMs per server, optimized virtualization features such as AMD Virtualization^TM (AMD-V^TM) technology, and unique energy efficiency features (AMD-P), Six-Core AMD Opteron^TM 8400 series processors are ideal for these applications.

When comparing the feature sets of Six-Core AMD Opteron^TM 8400 series processors and Intel Xeon 7400 series (“Dunnington”) processors, the feature set consistency of the AMD Opteron^TM 8400 series processors is in stark contrast to the Intel Xeon 7400 series offerings. Four items about the Xeon 7400 series stand out:

• The feature set differs dramatically between each of the seven models in the Xeon 7400 series
• No Xeon 7400 series processor includes RVI or Tagged TLB virtualization features
• The low-power Xeon L7445 and L7455 do NOT include the same energy efficiency features as the highest performance Xeon X7460
• All Xeon 7400 series processors use Fully Buffered DIMM (FB-DIMM) memory

The feature set consistency of the AMD Opteron^TM 8400 series processors gives customers a more predictable approach to data center planning and helps capacity planning, software image development, and validation efforts.

What about the performance and value of servers using these processors? If servers using Intel processors are clearly superior, then the differing feature sets won’t matter, right?

While the 46% VMmark performance advantage that a four-socket AMD Opteron^TM 8400 series processor-based server has over the top-performing four-socket Intel Xeon 7400 series processor-based server is impressive, I think that the performance of energy efficient Six-Core AMD Opteron^TM 8425 HE processor-based servers is even more impressive. This comparison shows that, depending on benchmark, a server using low-power Six-Core AMD Opteron^TM processors can provide nearly double the server performance of a server using low-power Hex-Core Intel Xeon L7455 processors at nearly half the processor price. Servers using low-power Six-Core AMD Opteron^TM 8425 HE processors even significantly outperform servers using the highest performance, highest power Hex Core Intel Xeon X7460 processors, again at nearly half the processor price.

When combined, all of these advantages (superior performance, energy efficiency, virtualization features, and pricing) make the Six-Core AMD Opteron 8400 series processors the ideal solution for four-socket servers.

Now, let’s go back to my original analogy about the hybrid SUV. In theory, the reason that people buy an SUV is because they need to move a lot of stuff, whether that’s people, groceries, or furniture. It’s the same way with four-socket servers – people buy a four-socket server because they need lots of memory to process lots of stuff, whether that’s database applications, virtualization/consolidation, or processing large datasets in high performance computing applications. Why shouldn’t you able to process lots of “stuff” and get energy efficiency too?

Andy Parma is a Product Marketing Manager for Server/Workstation products at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

In our booth on the exhibition floor we demonstrated the ability to use VMware’s VMotion to seamlessly migrate a running virtual machine  between an HP Proliant G2 server running Dual-Core AMD Opteron “Rev F” processors (released in 2006), G5 server running Quad-Core AMD Opteron “Shanghai” processors, (released in 2007) and the newest G6 server with Six-Core AMD Opteron processors.  We’ve just released a video on YouTube featuring a similar demo -- except that it also includes our upcoming processor codenamed “Magny-Cours”.

www.youtube.com/watch?v=kEmqz0SYmTg

I’m constantly getting questions on the topic of live migration and VMotion in particular.  Customers want to be assured that they can use VMotion to migrate virtual machines between older and newer versions of their AMD processer-based platforms.   Apparently we and our technology partners could do a better job of getting the word out. 

So let’s be clear -- the answer is a resounding YES!

Here are the facts.  With the launch of our first quad-core processors, AMD formalized a functionality that had existed in our processors for a while under the feature name of Extended Migration.  This capability enabled virtualization software vendors like VMware to support the live migration of a virtual machine between different versions of AMD processors starting with Rev E, which was released in 2004.  Subsequently, with ESX 3.5 u2, VMware released a feature as part of VMotion called Enhanced VMotion Compatibility (EVC) that uses AMD Extended Migration to support VMotion between AMD Opteron processor generations.  In order to utilize the feature, each server must be configured for EVC.  For information on configuring EVC, see the “Migrating Virtual Machines” chapter in the Basic System Administration Guide for your ESX release.  You can also gain a better understanding of EVC processor support and baseline definitions at the Enhanced VMotion Compatibility (EVC) processor support VMware web page. 

Keep in mind that live migration across AMD Opteron^TM processor generations is supported by other virtualization software, including Citrix XenMotion and Microsoft HyperV Server 2008 R2 Live Migration. In fact, you can watch a demo showcasing Hyper-V live migration between Quad-Core and Six Core AMD Opteron processors.

And don’t forget the technology demonstration we did a year ago with Red Hat, showcasing live migration between an AMD and Intel processor -- the YouTube video is still getting a lot of views.

Rest assured, AMD is pushing the technology envelope to cover all bases when it comes to offering broad support for your virtualization environment.

Tim Mueting is a Product Marketing Manager at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

Since the question was raised – I have reconfigured the VMmark systems as of May 6, 2009. I have listed valid memory configurations for the systems that match the total amount of memory used in the VMmark benchmark.

 * All cost comparisons are based on the difference in total system cost of the Intel processor-based system corresponding to the VMmark scores noted above, compared to the total cost for the AMD Opteron™ processor-based system referred to in the corresponding section of the above chart.  Prices are based on configurations submitted on OEM (Dell and HP) online system configuration tools as of May 7, 2009. 

 Now instead of focusing on who wins – AMD or Intel – let’s focus on details that are probably more important to customers who are really using these systems. As is shown in the chart above, with systems based on the AMD Opteron™ processor Model 2384, there are more options of memory speeds and DIMM configurations when purchasing 64GB of memory. This provides the ability to choose in terms of system price or performance.  I think this kind of choice puts the customer in the winner seat.

For another look at system configurations and pricing I suggest you go to the Solutions Oriented Blog. 

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

 As described by VMware, Fault Tolerance helps provide continuous availability with a simple ‘click of a button’ – something that gives businesses with virtualized data center environments peace of mind. At AMD, we have been thinking about continuous availability and disaster recovery for years and architected our AMD Opteron^TM processors, in part, to address this. If you are a regular reader of my blog, you may know we’ve focused this effort on the hardware side with our live migration capabilities (something we specifically call Extended Migration) across all of our processor generations, dating back to single-core Opteron processors! But why is this important? Well, with an AMD Opteron processor-based data center, an IT manager can dynamically move VMs across physical servers and in the case of a hardware failure-do so without interrupting services to users. VMware takes this one step further with Fault Tolerance software that is designed to provide even more reliability.

 What is so interesting about VMware’s Fault Tolerance is that it works to create a “shadow instance” of the virtual machine that is in lockstep with the primary instance (not surprisingly, part of VMware’s vLockstep technology). We have worked with VMware to make sure our hardware provides a solid foundation for their technology, and are extremely excited at the value it can offer our customers. More than anything, it addresses the number one concern we hear from customers when they consider virtualization: how reliable is this technology? What happens if there is an earthquake or a pipe bursts in my office? How do I ensure my data is protected?

 Of course, it’s natural to be a bit hesitant at the thought of virtualizing your most critical business applications, but it’s my job to let customers know these risks are addressed with virtualization. Unique hardware features like AMD Extended Migration and software like VMware’s Fault Tolerance not only help your trusty IT manager (and you) sleep better at night, but also help make their job easier by including reliability features directly on the hardware and software. Its complex technology made simple. I like the sound of that. 

is a Product Marketing Director at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

Margaret Lewis, director, AMD commercial software and solutions

Margaret Lewis

@margaretjlewis

Margaret Lewis, director, AMD commercial software and solutions

Margaret Lewis (@margaretjlewis) is a Product Marketing Director at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

I’m John Troyer from VMware’s VMTN Blog. For AMD’s Virtualization Ecosystem Month, I thought I’d talk about the recent performance testing we’ve seen with the 45nm Quad-Core AMD Opteron™ processor (codenamed “Shanghai”) and how VMware Virtual Infrastructure can unleash your applications to take full advantage of your new hardware.

First of all, these processors are fast. Here is some background on VMmark from our VROOM! Blog:

VMmark is a benchmark intended to measure the performance of virtualization environments in an effort to allow customers to compare platforms. It is also useful in studying the effect of architectural features. VMmark consists of six workloads (Web, File, Database, Java, Mail and Standby servers). Multiple sets of workloads (tiles) can be added to scale the benchmark load to match the underlying hardware resources. For more information on VMmark see here.

Now take a look at the VMmark Results page: a screen shot of the three current top scoring entries is shown on the right. The Quad-Core AMD Opteron processor Model 8384 is powering all of those top-scoring entries.

One of the reasons these Quad-Core AMD Opteron processors are so fast is that they contain AMD’s Rapid Virtualization Indexing (RVI) technology, which gives hardware support for MMU virtualization. We wrote a white paper showing how VMware ESX takes advantage of RVI, including a 29% increase in performance with Citrix XenApp; and in a recent study on our VROOM! blog, we showed RVI boosting VMmark performance boost by as much as 17%.

These processors are fast, but can your application take advantage of them? Many applications can’t utilize all the cores of modern CPUs. We tested the ability of the Apache web server to scale up and use more CPU cores using the web performance benchmark SPECweb®2005. Even using 8 processors, .However, if you scale out with VMware Virtual Infrastructure using multiple virtual machines, all running a web server, you can scale to take advantage of all 8 cores.Apache on a native, non-virtualization system only delivered 1.85x the performance of a single processor

We decided to show this by creating a rocket car and setting a land speed record. We documented how we built our web serving rocket car on an HP ProLiant DL585 G5 with four Quad-Core AMD Opteron Model 8382 processors. This setup achieved the highest SPEC®web2005 result ever on a 16-core system. Now, there are cheaper ways to move this many transactions on a virtual platform, but we weren’t building a cargo-hauling truck, we were building a race car to prove a point – scaling out with virtual machine building blocks is the best way to overcome multi-core scaling limitations of current applications.

Check out the eye-opening graphs in Scott Drummond’s blog post showing how your application running in a native environment probably will (or won’t) scale over the next few years and how a virtual building block architecture can help you take advantage of the newest generation of processors. VMware will continue to work together with AMD to help unleash the power of your high performance applications with virtualization.

John Troyer manages VMware’s blogging and social media programs.His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Any claims made herein have not been independently verified by AMD. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

SPEC and SPECweb are registered trademarks of the Standard Performance Evaluation Corporation.