VIRTUALBOY BLOG
Anyone who’s deployed Hyper-V as part of either Windows Server 2008 R2 Core Install, or, through Hyper-V Server 2008 R2, will know that changing certain settings requires a bit of work in the command line environment. For some people, easy peasy, but for others, it was a step too far, and back to the GUI they went. Then came Core Configurator, which bridged the gap between command line and GUI, without sacrificing reduced footprint, patching, services overhead etc.
Even with Core Configurator, there were still some bits that couldn’t be performed. Take Network Adaptor Protocols. How can you turn them off, turn them on, make changes, change orders etc? If you’re not sure what I’m referring to, I effectively mean:
How do you get these tick boxes from a command line?
John Howard, Senior Program Manager in the Hyper-V team, has blogged about a tool, developed by our very own Keith Mange, called NVSPbind, which gives users easier, granular control over network items they want enabled for their physical NICs and helps fills a gap in troubleshooting, optimising and hardening networking configurations on Server Core or Hyper-V Server. You can get all the info here.
As John says, with all utilities which change network configurations, you have to be extremely careful as you may disrupt or even lose network connectivity if you are managing a machine remotely. It may be handy to have Keith's other utility around just in case. http://code.msdn.microsoft.com/nvspscrub. NVSPscrub is a tool for removing Hyper-V Virtual Networking Configuration from the parent partition. It is especially useful in Server Core environments when the Hyper-V UI may not be available, and if you’ve fiddled with NVSPbind, and things have taken a turn for the worse!
This is a combination of technology that I’m starting to see on a pretty regular basis. People looking at virtualisation, are typically weighing up new hardware, and from a server perspective, it makes sense to go for the latest and greatest CPUs (from both Intel and AMD) because of their virtualisation optimisations. When it comes to storage however, more and more customers are learning to appreciate the performance that iSCSI can bring, yet at a great price point when compared with Fibre implementations. This isn’t obviously correct for all scenarios, but, typically the rule holds. Stephen’s got a great example of this here. Stephen’s also got a link to a test performed back in March last year where Microsoft and Intel, combining the technologies above, pushed over 1 million IOPS.
If you are thinking of going down this route, combining these technologies, then I’d advise you to have a look at recent information produced by Microsoft and Intel around this very combination. The stuff has been floating around the web on various blogs, news sites and so on, for a good few while now, but today was the first time I’ve had chance to have a look in more detail, and the results are very encouraging indeed.
In terms of resource and information, firstly, let’s get the marketing document out of the way. This brief highlights the key benefits when you go down the ‘combination’ yet doesn’t go into the nuts and bolts like some of the other resources do. Among other things, it highlights technologies like Microsoft’s VMq and Intel’s VMDq and how this combination can accelerate network traffic, including iSCSI. It’s a useful starting point though.
Going back to what I said earlier about the 1 million IOPS, if you look at Stephen’s post, one of the comments is around the fact that it’s 1 million IOPS with 512 byte blocks, which isn’t reflective of applications like Exchange and SQL, which have larger recommended block sizes, and as you increase the block size, the number of IOPS will decrease. This isn’t to say the test is flawed, nor is it marketed to hit the headlines. What it proves is, iSCSI is ready for the big time, provided it is specced accordingly, and when I say that, don’t instantly think “This must be the highest-end SAN, and the highest-end iSCSI HBA’s” to achieve anything near those levels – if you read Stephen’s post carefully, you’ll find that the test was run using the in-box Windows Server 2008 R2 iSCSI initiator, on an Intel 5500 (which ships as standard in most boxes now), and it ran over a regular 10GbE Intel Ethernet NIC. Sure, 10GbE is expensive, compared with 1GbE, but if you need that level of performance, achieving it with iSCSI is a very valid alternative to the Fibre approach and gives you more options around passing iSCSI traffic into VMs than Fibre will.
To quote Stephen
Performance is not an issue for iSCSI – Sure, not every iSCSI stack can handle a million IOPS, but the protocol is not the problem. iSCSI can saturate a 10 GbE link and deliver all the IOPS you might need. Performance is not an issue for software – Today’s CPUs are crazy fast, and optimized software like the Windows Server 2008 R2 TCP/IP and iSCSI stacks can match or exceed the performance of specialized offload hardware. Storage vendors need to step up their game – Whose storage array can service a million iSCSI IOPS? Raise your hands, please! I can’t hear you! Hello? Anyone there? Fibre Channel and FCoE don’t rule performance – I don’t know of a Fibre Channel SAN that can push this kind of throughput or IOPS through a single link. Even FCoE over the same 10 GbE cable can’t quite do it. If they are to stay relevant, they had better come up with a compelling advantage over iSCSI!
Performance is not an issue for iSCSI – Sure, not every iSCSI stack can handle a million IOPS, but the protocol is not the problem. iSCSI can saturate a 10 GbE link and deliver all the IOPS you might need.
Performance is not an issue for software – Today’s CPUs are crazy fast, and optimized software like the Windows Server 2008 R2 TCP/IP and iSCSI stacks can match or exceed the performance of specialized offload hardware.
Storage vendors need to step up their game – Whose storage array can service a million iSCSI IOPS? Raise your hands, please! I can’t hear you! Hello? Anyone there?
Fibre Channel and FCoE don’t rule performance – I don’t know of a Fibre Channel SAN that can push this kind of throughput or IOPS through a single link. Even FCoE over the same 10 GbE cable can’t quite do it. If they are to stay relevant, they had better come up with a compelling advantage over iSCSI!
What you’ll also find from Stephen’s post, is that Hyper-V also performed admirably, with performance remaining similar to native throughout.
If you can, watch this webcast. It’s available on demand, and presents to you, first hand, the results that Stephen has referenced above.
Hat-tip to the RDS Team Blog for this one.
One of the biggest requests I’m hearing around RDS in 2008 R2 and Windows 7, is for RDP performance information and benchmarks. For those interested, it’s a close call on requests for this information, or for RDS Sizing guidance. The sizing guidance is coming soon, trust me!
So, back to the RDP Performance Improvements, to quote the RDS Team…
“As with Remote Desktop Services in Windows Server 2008 R2, virtual machine-based desktop virtualization faces increasing performance challenges when enterprises attempt to use this technology to support a globally distributed workforce. A key consideration of performance relates to Remote Desktop protocol efficiency which continues to present an issue for bandwidth constrained environments. This limitation can manifest itself by limiting the number of users who can access virtualized desktops (user density) over available bandwidth, and with a degraded user experience. Remote Desktop Protocol (RDP) 7.0, similar to previous RDP versions, provides a competitive experience for low bandwidth (e.g. 56 Kbps) connections. After bandwidth requirements, network latency is the second fundamental challenge for customers and partners that wish to deploy virtualized desktops for a broad range of end-users and applications”
Very true indeed. If you’d have said, 18 months ago, you can play a rich multimedia video over RDP, and it wouldn’t be jerky, out of sync etc, people would have laughed at you, yet that laughter would have been mostly based on older iterations of RDP, and it’s traditional, functional use within an infrastructure. Fast forward to now, with RDP7, that kind of experience is a reality, but would I expect that level of rich experience over all connectivity types, fast and slow? No, inevitably not. As you’d expect, the experience would degrade, especially over the WAN, which could, as the paragraph above suggests, degrade the experience for others too. Therefore, understanding the RDP protocol in the latest release is pivotal when evaluating an RDS infrastructure.
So, back to the whitepaper…
“With the release of the Windows Server 2008 R2 and Windows 7 operating systems, Remote Desktop Protocol (RDP) is more feature rich, enabling new presentation and remote-oriented functionality such as accelerated bitmap rendering, multi-media redirection streaming, and network topology awareness. As these features become integrated in the enterprise environment, it is important to analyze and understand their impact on your current network infrastructure and the end-user experience. This paper details the various RDP features and the potential improvements to usability and quality of the end-user remoting experience, as well as system deployment metrics. To test the impact of different features and compare RDP 7.0 to the previous RDP 6.1 version, we performed a variety of tests by using automated and simulation tools to demonstrate the user scenarios outlined in this white paper. These tests are broken down into two broad groups: a set that simulated a user working with actual Microsoft® Office applications at realistic speeds, and a set of multi-media scenarios that simulated a rich media environment commonly expected by today’s and tomorrow’s users. To test the impact of different features and compare RDP 7.0 to the previous RDP 6.1 version, we performed a variety of tests by using automated and simulation tools to demonstrate the user scenarios outlined in this white paper. These tests are broken down into two broad groups: a set that simulated a user working with actual Microsoft® Office applications at realistic speeds, and a set of multi-media scenarios that simulated a rich media environment commonly expected by today’s and tomorrow’s users. In addition, some of the scenarios have been run at increased network latencies to demonstrate the viability of using RDP in distributed environments, such as in Branch offices or telecommuting scenarios where broadband capabilities exist”
“With the release of the Windows Server 2008 R2 and Windows 7 operating systems, Remote Desktop Protocol (RDP) is more feature rich, enabling new presentation and remote-oriented functionality such as accelerated bitmap rendering, multi-media redirection streaming, and network topology awareness. As these features become integrated in the enterprise environment, it is important to analyze and understand their impact on your current network infrastructure and the end-user experience. This paper details the various RDP features and the potential improvements to usability and quality of the end-user remoting experience, as well as system deployment metrics. To test the impact of different features and compare RDP 7.0 to the previous RDP 6.1 version, we performed a variety of tests by using automated and simulation tools to demonstrate the user scenarios outlined in this white paper. These tests are broken down into two broad groups: a set that simulated a user working with actual Microsoft® Office applications at realistic speeds, and a set of multi-media scenarios that simulated a rich media environment commonly expected by today’s and tomorrow’s users.
To test the impact of different features and compare RDP 7.0 to the previous RDP 6.1 version, we performed a variety of tests by using automated and simulation tools to demonstrate the user scenarios outlined in this white paper. These tests are broken down into two broad groups: a set that simulated a user working with actual Microsoft® Office applications at realistic speeds, and a set of multi-media scenarios that simulated a rich media environment commonly expected by today’s and tomorrow’s users. In addition, some of the scenarios have been run at increased network latencies to demonstrate the viability of using RDP in distributed environments, such as in Branch offices or telecommuting scenarios where broadband capabilities exist”
There’s actually some very useful information in the whitepaper, particularly around what tweaks you could make to your user experience to optimise the bandwidth utilisation. One thing I found useful, is the frame-rate achieved with Silverlight content, versus Flash, at different colour levels:
I really didn’t know that!
This, and other useful bits of info, are available in the whitepaper. Check it out now!
First look of Windows Phone 7 for me, and I' have to say, I’m impressed!
Very Zune HD-like. Very nice indeed. Hopefully the community will rally around Windows Phone 7 and build some half decent apps for it!
Anyone who’s seen my present Microsoft’s Virtualisation & Management story, will know I spend quite a bit of time focussing on System Center Operations Manager. For me, I think it’s the mission control of the System Center Suite, where, typically, IT Admins’ would spend the majority of their time. Whether it’s proactive monitoring and alerting, or reporting through Service Level Dashboards, Operations Manager gives us excellent insight into the environment, whether it’s physical, or virtual.
How does Operations Manager get it’s knowledge about Technology-X?
Simple. Management Packs. Important to note, MP’s are not written by the Operations Manager team. They are written and developed by the team (Microsoft or 3rd Party) that makes that respective technology. Exchange MP is written by the Exchange team, SharePoint MP by the SharePoint team, and so on.
Where do you get these Management Packs?
Well, with Operations Manager 2007 R2, you can search and import MP’s right into the Operations Manager interface, however, those who’ve used Operations Manager for some time, will know you have to visit the MP Catalogue. If you’ve not used the MP Catalogue for a while, the first thing you’ll notice is that the site has been moved into the PinPoint set of sites, and in my view, isn’t quite as easy to find the MPs as before, but I’m sure it’ll get better with time. Don’t shoot the messenger on that one!
New Management Packs?
Well, a few highlights for me include:
Operations Manager Management Pack for Forefront Threat Management Gateway (TMG) 2010 (Download)
This MP monitors Forefront TMG and includes monitors and rules to track the deployed topology & features, performance, availability, and reliability of Forefront TMG components. With detailed alert information, you can quickly identify and troubleshoot Forefront TMG issues, minimizing time-to-resolution when problems occur. You can collect and analyze performance trends and metrics, and obtain performance information that allows you to manage bottlenecks, identify capacity requirements, and proactively manage your Forefront TMG deployment to resolve issues before problems occur.
Operations Manager Management Pack for Windows 2008 R2 Direct Access Server (Download)
Through Operations Manager, the Direct Access Server Management Pack supports the following features:
Automatic discovery of the Direct Access Server and its components, including:
Monitors that identify:
Operations Manager Management Pack Microsoft Exchange Server 2010 (Download)
The Microsoft Exchange Server 2010 management pack is designed to be used for monitoring Exchange 2010 events, collecting Exchange component-specific performance counters in one central location, and for raising alerts for operator intervention as necessary. By detecting, sending alerts, and automatically correlating critical events, this management pack helps indicate, correct, and prevent possible service outages or configuration problems, allowing you to proactively manage Exchange servers and identify issues before they become critical. The management pack monitors and provides alerts for automatic notification of events indicating service outages, performance degradation, and health monitoring.
The Exchange Server 2010 Management Pack includes rules and scripts to monitor and report on performance, availability, and reliability of all Exchange 2010 server roles including.
Operations Manager Management Pack for Forefront Protection 2010 for Exchange Server (Download)
The Management Pack for Forefront Protection 2010 for Exchange Server (FPE) allows you to discover FPE installations and components and to monitor them within System Center Operations Manager 2007. When there is an issue that may impact the availability, configuration, or security of your FPE deployment, Operations Manager uses the management pack to detect the issue, alert you to its existence, and facilitate diagnosis and corrective actions.
Operations Manager Management Pack for SharePoint 2010 Beta (Download)
This System Center Management Pack is for Microsoft SharePoint 2010 Products Beta, which includes monitoring for:
Operations Manager Management Pack for Windows Power Management (Download)
The Power Management Pack for Operations Manager 2007 R2 enables you to monitor and manage the power consumption of computers running Windows Server 2008 R2.
This management pack provides:
These are just a few of the MP’s available, however, don’t just import them all and expect everything to be perfect. Importing of MP’s should be phased, otherwise, as expected, you’ll get a lot of alerts, as the thresholds won’t be ideal for your environment in every case. So, import, tweak and tailor, a few at a time, and what you’ll end up with is a more tailored, more relevant, monitoring environment.
These, and all other MPs, can be found on the Management Pack website (or downloaded with Operations Manager 2007 R2).
Hat-tip to Ben for this one.
If you’re interested in the kind of things that can impact performance within Virtual Machines, from a networking perspective, and how you can optimise accordingly, this article may be of use to you. In a nutshell, increasing the VMBus buffer size within the VM, can increase network throughput to guest VMs, in some cases, very significantly indeed. Obviously it relates to your environment, and the workloads generating the network loads, but it’s definitely worth a read.
There are a number of useful suggestions in the article, along with how to make the changes.
That’s the networking done, but don’t forget about the Performance and Tuning Guidelines for Windows Server 2008 R2 that I told you about a week or so back too. All useful stuff!
For those of you interested in backup, and in particular, System Center Data Protection Manager 2010, the guys over at ZDNet Australia have written an interesting review of the beta release (Release Candidate coming soon!).
The review focuses on a few key areas, such as general usability improvements over DPM 2007, scalability improvements, automation improvements, client protection, protection around Virtualisation, SharePoint and Exchange 2010, along with improvements in DPM chaining and non-domain protection (great for hosting/managed services).
You can read the whole review here.
The thing that stood out most to me, was the bolt-on technology appliance, namely the EVault Backup and Recovery Appliance, produced by i365, which helps DPM 2010 become a more-heterogeneous protection tool, protecting UNIX, Linux, Novell Netware, VMware, IBM System i, and Oracle. It also provides the capability to protect up into the cloud, and it’s powered by Seagate. Useful stuff, I’d say.
Read all about that, here.
You just couldn’t write this script! Yesterday, I blogged about a new whitepaper which discusses the performance improvements of RDP7 over previous versions, and the kinds of factors you can expect to encounter that can impact the user experience using RDP7 and RDS. In that blog post, I also noted the number of requests for sizing guidance around the solution, specifically stating it was coming soon. Well, I didn’t think it would be this soon!
“The Remote Desktop Session Host (RD Session Host) role service lets multiple concurrent users run Windows-based applications on a remote computer running Windows Server 2008 R2. This white paper is intended as a guide for capacity planning of RD Session Host in Windows Server 2008 R2. It describes the most relevant factors that influence the capacity of a given deployment, methodologies to evaluate capacity for specific deployments, and a set of experimental results for different combinations of usage scenarios and hardware configurations.”
Definitely worth a read if you’re thinking about sizing RDS, including if you’re thinking of sizing it up on a Hyper-V platform. You can grab the document, here.
If you’ve been following my coverage of DPM 2010 over the last few months (here, and here), you’ll be pleased to learn that the Release Candidate build has been declared, and released to the web, on the Microsoft Connect website.
If you’re interested in downloading and evaluating this release (which I’d strongly recommend!), you can grab all the bits here.
If you have a look around the web (using your favourite Bing search engine), for virtualising/virtualizing SAP, you’ll find a lot of links for virtualising that workload on VMware, but what about Hyper-V? Do SAP support Hyper-V as a platform? Are there any recommendations if they do? Yes, and yes, and with the latest announcement, they now support Hyper-V R2 as the virtualisation platform, which brings in significant advantages around Live Migration, Second Level Address Translation (SLAT) support, and improved networking capabilities to name but a few.
Last year, there was a best practices white paper released, which detailed optimum configurations for SAP virtualised on R1 of Hyper-V, so this is still going to be relatively useful for R2, especially when combined with the Performance Tuning Guidelines for Windows Server 2008 R2 that I blogged about a few weeks back.
If you head on over to the SAP Community, and look at the Virtualisation on Windows page, you’ll see the statement of support front and centre, however one thing to be aware of, is that SAP will support Hyper-V R2 as the platform, but the guest OS that SAP runs on has to be Windows Server 2008, or 2003.
If you’re interested, there’s more guidance here:
If you’re thinking about virtualising SAP, ESX isn’t the only hypervisor in town…