Best of Microsoft Management Summit 2011 (#mmsuk2011)

A couple of weeks ago, I spent a day at Microsoft’s Best of MMS 2011 event in London – reacquainting myself with the latest developments in System Center. It was a pretty full day (and a pretty full venue – Microsoft’s London offices are far from ideal for this type of event, especially when the foyer is filled with partner booths) and there were plenty of demonstrations of product features and advantages (although, in true software vendor style, not too much focus on business benefits).

This post brings together my notes from the event, picking up the highlights from the keynote, supplemented with a few more from the individual product sessions:

  • Consumerisation is not just about devices but also management and security.
  • System Center Configuration Manager (SCCM) 2012 is about empowering users – no longer device centric but user centric – application delivery is context sensitive to the device that the user is using at that time. SCCM 2012 includes mobile device management: managing settings and policies for any device that can use Exchange ActiveSync
  • Forefront Endpoint Protection (FEP) is now using SCCM – so it’s no longer necessary to have separate infrastructures for management and security – also FEP is now part of the core CAL (as is the Lync standard CAL). New 2012 release of FEP will run on SCCM 2012 (currently it runs on SCCM 2007 R2/3).
  • Windows Intune is a cloud-based solution for light management/unmanaged PCs (no on premise infrastructure required). It includes software assurance for Windows Enterprise (so users can stay on the latest Windows release).
  • There are various marketing pitches about the cloud – but it’s really a model for computing and not a place/destination. Cloud attributes include self-service, shared (there may be some logical partitioning), scalable/elastic, usage-based chargeback.
  • IT as a service includes: IaaS (addition of infrastructure resilience); PaaS (not worried about virtual machines); SaaS (consuming an application directly from a vendor).
  • Microsoft’s own datacentre infrastructure is based on extreme standardisation; business alignment (service-specific characteristics); SLA-driven architecture; and process maturity (re-imagined processes – not just automating today’s processes but thinking about the most efficient process for tomorrow, automation, change control).
  • Private cloud is a combination of virtualisation and management – adopting public cloud practices internally… it’s not just about virtual machines and other infrastructure – it’s a full stack of management capabilities.
  • The Microsoft stack is optimised for Microsoft software but there are also some cross-platform capabilities in System Center Operations Manager (SCOM) and in System Center Virtual Machine Manager (SCVMM).
  • Cloud services (public or private) are based on a provider-consumer relationship. A typical service provider role might be a data centre administrator, whose concerns would be fabric assembly(storage/network/compute), delegation and control, flexibility and elasticity, and cost efficiency. A consumer example is an application owner, who is looking for empowerment and agility, a self-service experience, application visibility and control, and simplicity.
  • System Center codename Concero is a new (web-based) product in development for (cloud) application owners, providing a view of all public and private clouds (Windows Azure subscriptions and on premise infrastructure, not just Hyper-V but ESX and Xen too). Pick a template and build out components (different tiers) for services within existing clouds. Configure the attributes that an application owner has to manage. Not just virtual machines but other data centre hardware (load balancers, etc.) too, using SCVMM in the background to deploy.
  • Request a new cloud using a catalogue from System Center Service Manager (SCSM).
  • Delivering a private cloud is about creating logical and standardised structures (because there is lots of legacy to manage there will always be a diverse infrastructure) and delegating portions to business functions.
  • SCVMM 2012 supports creation of delegated private cloud infrastructure – create a logical cloud by defining attributes such as number of virtual machines, hypervisor choice, available service templates, and what can be done with these resources.
  • Applications need to be abstracted from infrastructure (externalised configurations).
  • Business empowerment is not about virtual machines though (SLA management – and self service too) – SCOM 2012 and Avicode (recent acquisition) give application insight to create dashboards for cloud applications and and drill down into alerts. These dashboards may be made available to managers via SharePoint web parts. SCOM 2012 also includes network monitoring.
  • System Center Orchestrator (SCO) is the new name for Opalis (process automation tool) providing run books automating operational processes.

System Center Roadmap 2011

Some SCVMM highlights:

  • SCVMM is now about far more than just virtual machines (I wonder when it will be renamed – perhaps System Center Fabric Manager?). Enhancements include:
    • Infrastructure (high availability/cluster aware, easier upgrade path, custom properties with name/value pairs, fully scriptable via PowerShell).
    • Fabric management (bare metal provisioning of Hyper-V using Windows Deployment Services and host profiles, multiple hypervisor support – Hyper-V/ESX/Xen, network management and logical modelling, storage management using standards such as SMIS, update management, dynamic optimisation, power management/smart shutdown – integrated with baseboard management controllers, cluster management).
    • Cloud management (application owner usage, capacity and capability, delegation and quota).
    • Service management (service templates, application deployment, custom command execution, image based servicing).
  • SCVMM works with SCOM for load balancing (uses a connector and rebalances when limits hit which is a reactive approach) – in 2012 it also allows proactive load balancing (dynamic optimisation). This can also be used to schedule host power-downs.
  • Self-service portal is integrated in SCVMM 2012. Console is now context-aware so it can be used by all user roles and they only see delegated resources.
  • Server App-V is part of SCVMM 2012 – separating app state from the operating system, to enable image-based servicing and slide in a new operating system instead of traditional operating system updates. It is intended for line of business apps, not SQL, Exchange, SharePoint, etc.
  • Service designer to create 3 tier applications and template them within the VMM library. Define deployment order and how to scale out. Scale out a virtual machine tier via a right click (within the service definition) – Microsoft also plans to deliver a management pack to detect service performance from SCOM and scale accordingly.
  • Roles and features now part of operating system configuration in VM templates, as are application configuration items – not just virtualised but also with scripts. Deploy service and will be intelligently placed.
  • Still support a server-based approach but trying to bring a service-based approach to deploying and managing apps in DC. This is also represented in SCOM.

More on SCO:

  • In the private cloud change happens all the time but it’s the same change each time – so management is not about approval but logging. We can remove the manual but do need the ability to chose (to cope with diversity, move at different speeds).
  • Three step process:
    • Integrate - take things (like disparate System Center products) and reference them as single entity.
    • Orchestrate - make them work together.
    • Automate - make things happen automatically
  • If we jump straight to automation, we haven’t re-imagined the process. That means that if we take a bad process and automate it, we get a fast bad process! And if that breaks things, it really breaks them!
  • SCO (Opalis) concepts include:
    • Activities – intelligent tasks with defined actions.
    • Integration packs – extendable connectors to communicate with other solutions (outbound – SCO has an application integration engine in a web service form for inbound communications)
    • Databus – publish and consume mechanism (when something happens, capture information, put it on the bus, send along as it works through the runbook).
    • Runbooks – system level workflows that execute a series of linked activities to complete a defined set of actions.
  • SCO behaves in the same way as Opalis 6.3 with some minor UI changes and some investments in functionality but no fundamental changes in the way the product works (although it will be available in additional languages). It is a 64-bit only product.
  • SCSM also has an orchestration engine that is not based on Opalis – this will remain as a separate but complementary product.
  • Some integration packs have been remediated for SCO and will be available out of the box but not all – the packs remaining are not tied to service packs, etc. and will be released out of band.

The next session was, frankly, dull, droning on about SCSM and “GRC”, and I missed the presenter introducing the term (which I now know is governance risk and compliance – it was on the title slide of the deck but there was no definition). I have no notes to share as I struggled to keep up from the start…

Moving on to System Center Data Protection Manager (SCDPM):

  • SCDPM 2006 provided centralised file-based backup (removing tapes from branches).
  • DPM 2007 included Volume Shadow Copy Service (VSS) application support for Exchange, SQL Server and SharePoint.
  • DPM 2010 included more enterprise features and client support. It still requires Active Directory but can now backup standalone machines off domain. Supported applications include Exchange, SQL Server, SharePoint, Dynamics, Virtual Server/Hyper-V, Windows Servers and Clients – and it can be used to backup SCDPM too. It can also backup highly available configurations. 
  • SCDPM can create backups every 15 minutes with one full backup, then block-level differentials. Online snapshots for disk-based recovery and tape-based backups. Initial backup can be immediate, scheduled or via removable media.
  • Many application owners use SCDPM and consider it as an extension to their application, rather than as a backup tool.
  • SCDPM 2012 introduces a centralised management  for up to 100 SCDPM 2010/2012 servers or 50,000 sources; role based management; push to resume backups; SLA-based reporting (don’t alert every failure, just those that matter); consolidated alerts (fix one problem, not 20 alerts) and extensibility via PowerShell to script known issues.
  • The console uses SCOM (it is a management pack and a few binaries) and may be integrated with a ticketing system (either SCSM or third parties products such as HP Openview via connectors). Roles are taken from SCOM (either create new roles or use existing ones).
  • Infrastructure enhancements include certificate-based authentication (where there is no NTLM trust in place) and smarter media co-location (choose specific data sources to share a tape).
  • Workload enhancements include SharePoint item level recovery, Hyper-V item level recovery (even when SCDPM is inside a virtual machine), and generic data source protection offering basic protection/recovery support for any referential data source with full application backup (full, delta and consistency check), original item recovery and restore as files to a network location, and XML support for applications without a VSS writer.
  • There is no native protection for non-Windows applications, but virtual machines with other operating systems (e.g. Linux) can be backed up – the key is VSS support.

Some of the key points I picked up from the SCCM presentation:

  • SCCM 2012 is less focused on packages and advertisements, now about applications, not scripts.
  • User-centric approach, with better support for virtual environments.
  • New models for communication between components.  Improved infrastructure architecture using SQL replication.
  • Now includes mobile device management capabilities that were previously in System Center Mobile Device Manager as well as support for “light” management of mobile devices via Exchange ActiveSync.

Finally, SCOM 2012:

  • Features simplified disaster recovery (for the SCOM servers) and monitoring improvements (so a device is monitored by a pool, not a single server).
  • Support for monitoring Linux machines was introduced in SCOM 2007 R2, 2012 includes network monitoring and application monitoring (.NET and Java, when running on Windows).
  • Network and application monitoring is not intended to be all-encompassing, but provides information to take to specialist teams and at least have an idea that there is an issue – more than just a gut feeling that the network or the application is “broken”.
  • Introduction of dashboard templates – web and console views – can publish link with others. Can also create more complex dashboards that can be integrated with SharePoint, customise data visualisations via widgets. Dashboards and widgets are delivered via management packs.
  • SCOM 2007 moved from server to service monitoring – 2012 is taking the next step.

As I look back on the day’s event I do have to congratulate the System Center product group on their openness (talking about future products in a way that helps customers and partners to plan ahead) and for running a free of charge event like this which is a great way for me to get the information I need, without the significant investment of time and money that conference attendance entails. Now, if only the Windows client and server teams would do something similar…

Slidedecks and recorded sessions from the Best of Microsoft Management Summit 2011 are available on the Microsoft website.

Microsoft Management Summit 2010 highlights

This week sees the annual Microsoft Management Summit (MMS) taking place in Las Vegas, with over 3500 attendees from around the world, even though there are many people stranded by the current flight restrictions in Europe.  According to Microsoft, that’s 50% up on last year – and those delegates have access to 120 break out sessions to learn about Microsoft’s vision and technology for IT management – across client devices, the datacentre and the cloud.

The keynote presentations are being streamed live but, for those who missed yesterday’s keynote (as I did) and who are waiting to hear today’s news, here are the main highlights from the event, as described by Paul Ross, a Group Product Marketing Manager for System Center and virtualisation at Microsoft.

Cloud computing is a major trend in the IT industry and many customers are trying to balance new models for elastic computing with trying to get the best TCO and ROI from their existing investments.  There are those who suggest Microsoft doesn’t have a cloud strategy but it’s now 5 years since Ray Ozzie’s Internet Service Disruption memo in which he set out Microsoft’s software plus services approach and Steve Ballmer reinforced Microsoft’s Cloud Services vision earlier this year.

For many years, Microsoft has talked about the Dynamic Systems Initiative (DSI), later known as Dynamic IT and the transition to cloud services is in line with this – model driven, service focused, unifying servers and management, thinking about services instead of servers, and automated management in place of manual approaches. Meanwhile, new deployment paradigms (e.g. virtualisation in the data centre) see customers shifting towards private and public cloud environments.  But customers are experiencing a gap in the consistency of security models and application development between on premise and cloud services – and Microsoft believes it is the key to allowing customers to bridge that gap and provide consistency of infrastructure across the various delivery models.

Some of the new products announced at this year’s MMS include the next version of System Center Virtual Machine Manager (SCVMM), slated for release in the second half of next year, and which will take a service centric approach to management – including new approaches to deploying applications. Alongside SCVMM, System Center Operation Manager (SCOM) will also be updated in the second half of 2011 – itself making the transition to a service-centric model.

Before then, June 2010 will see the release to web of the Dynamic Infrastructure Toolkit for System Center which provides enterprise customers with the foundations for creating a private cloud with concepts such as on demand/self-service provisioning, etc.

Today’s keynote will focus on the shift from device-centric computing to a user-centric approach.  Many organisations today operate separate infrastructures for different client access models – and there is a need for unification to manage IT according to end user requirements.  Central to this vision is the need to unify the products used for security and management of the infrastructure, reducing costs and focusing on user-centric client delivery for the cloud.

Earlier this week, we heard about the beta for Windows Intune – offering security, management, Windows Update and MDOP benefits within a single subscription for small to medium sized businesses.  Today’s headlines are enterprise-focused and will include the announcement of the beta for System Center Configuration Manager (SCCM) 2007 R3 – focused on power management and unified licensing for mobile devices alongside traditional desktop clients.  SCCM vNext (again, scheduled for the second half of 2011) will be focused on user-centric management – offering a seamless work experience regardless of whether applications are delivered via App-V, VDI, or using a traditional application delivery approach.  In addition, SCCM vNext will incorporate mobile device management (currently in a separate product – System Center Mobile Device Manager), allowing a single infrastructure to be provided (so, to summarise: that’s licensing changes in SCCM R3, followed by the technology the next release).

In other news, we heard yesterday about the release of System Center Service Manager (SCSM) 2010 and System Center Data Protection Manager (SCDPM) 2010 – both generally available from June 2010.  SCSM is Microsoft’s long-awaited service desk product – with 57 customers in production already and around 3000 on the beta – which Microsoft hopes will disrupt service desk market that they describe as being “relatively stale”.  Built as a platform for extension by partners SCSM includes the concept of process packs (analogous to the management packs in SCOM) and Microsoft themselves are looking to release beta compliance and risk process packs from June, helping to grow out the product capabilities to cover a variety of ITIL disciplines.  As for SCDPM, the product gains new enterprise capabilities including client protection (the ability to back up and recover connected client systems) – and both SCSD and SCDPM are included within the Enterprise CAL and Server Management Suite Enterprise licensing arrangements.

For some years now, Microsoft has been showing a growing strength in its IT management portfolio – and now that they are starting to embrace heterogeneous environments (e.g. Unix and Linux support in SCOM, ESX management from SCVMM), I believe that they will start to chip away at some of the territory currently occupied by “real” enterprise management products.  As for that image of a company that’s purely focused on Windows and Office running on a thick client desktop, whilst that’s still where the majority of its revenue comes from, Microsoft knows it needs to embrace cloud computing – and it’s not as far behind the curve as some may believe.  The cloud isn’t right for everyone – and very few enterprises will embrace it for 100% of their IT service provision - but, for those looking at a mixture of on-premise and cloud infrastructure, or at a blend of private and public cloud, Microsoft is in a strong position with a foot in either camp.

Microsoft System Center licensing gets a complex simplication treatment

Some time ago, I wrote that the most cost-effective way to license multiple System Center products is generally through the purchase of a System Center server management suite license, which includes licenses for System Center Operations Manager (SCOM), System Center Configuration Manager (SCCM), System Center Data Protection Manager (SCDPM) and System Center Virtual Machine Manager (SCVMM).

It’s worth noting that Microsoft made some changes this week which complicate things somewhat as, according to a communication that I received from a large account reseller (LAR):

“Effective on 1 July, 2009, with the release of Microsoft System Center Operations Manager 2007 R2, System Center Server Management Suite Enterprise (SMSE) will be switched from an unlimited operating system environment to a four operating system environment. A new suite offering, Microsoft System Center Server Management Suite Datacenter (SMSD) will be introduced and will include the same products as System Center Server Management Suite Enterprise, but, it will be licensed per processor and will provide for the management of an unlimited number of operating system environments.”

Whilst this kind of makes sense because it falls in line with the Windows Server virtualisation licensing it also has the potential to affect the cost of licensing management products in a virtualised environment as, where high levels of server consolidation may have previously been achieved and managed with an SMSE, now multiple SMSDs will be required.

Further information may be found on Microsoft’s How to Buy the System Center Server Management Suites page but there is also a Server Management Suite Editions FAQ for those who want to know the details.

Virtualised hardware hotel

I was at a VMware event yesterday where they proudly played this video…

…it’s a bit of fun (and the music is really catchy – even if the lip sync is a bit out!) and was apparently first shown at VMworld a few months back.

It’s not just VMware that can offer this type of solution though – I did use VMware Virtual Infrastructure (VI) in the design I produced for a server consolidation exercise with a “big four” accountancy firm a couple of years back but it was very expensive and required a huge leap of faith on the part of both the customer and the datacentre managed service provider. Now we’re in the second half of 2008, I’m not sure if I would be using VMware products in my “virtualised hardware hotel”. For a lot less money I could do the same thing with Windows Server 2008 and Hyper-V, together with System Center Virtual Machine Manager 2008. Some people will argue that the VMware products have maturity on their side and I’ll concede that it’s true – VMware did create the x86 virtualisation market – but a hypervisor (or virtualisation layer, in VMware-speak) is a commodity now and the simple fact is that I really can’t justify advising my clients to spend the extra money on ESX and Virtual Center, especially as the Microsoft offerings under the System Center banner can be used to manage my virtual and physical infrastructure as one.

If only Microsoft produced viral videos like this, I could share one with you… so come on Redmond… give me something to play back at the VMware boys (and girls).

Microsoft Licensing: Part 4 (System Center products)

Last week, I wrote about licensing Microsoft Server products but I deliberately ignored Microsoft’s family of systems management products. This post continues the series on Microsoft licensing, taking a look at the licensing considerations for the main System Center products.

System Center products that rely on SQL Server for database functionality, for example System Center Operations Manager (SCOM) 2007 and System Center Configuration Manager (SCCM) 2007 are available both with and without SQL Server 2005 standard edition included (which option to select will depend on the database arrangements in use). No SQL client access licenses (CALs) are required if the per-processor model is used, or if the inclusive SQL Server license is used; however SQL CALs are required for every managed device if SQL Server is licensed using the server and CAL model.  It’s also important to note that if the included SQL Server licensing is used, then SQL Server may only be used for System Center products – not as a standalone server or with any other application.

The main System Center products do not require CALs but a Management License (ML) is required for each managed device.For SCOM, there are some exceptions:

  • Devices that SCOM has merely discovered the presence of but for which SCOM is not being used for management.
  • Devices functioning only as network infrastructure devices (layer 3 and below).

Different MLs exist for client devices and servers with two server MLs available for SCOM – standard for monitoring basic workloads such as the operating system, networking, file and print services and management of the hardware, enterprise for other workloads (referred to by Microsoft as application and premium workloads).

In a virtualised environment, each operating system instance (OSE) is considered as a device and requires an ML. If the OSE is running a client operating system, then a client ML is required; if the OSE is running a server operating system, then a server ML is required.

SCCM follows similar rules, with a standard server ML being limited to operating system and basic workload desired configuration management, whilst an enterprise server ML is required for full application and server desired configuration management, including the proactive management of systems for configuration settings.

System Center Data Protection Manager (SCDPM) 2007 also has two types of server ML – standard for recovery and backup management of file servers and enterprise for applications including SQL Server, Exchange Server, and Office SharePoint Server. The Enterprise server ML also includes the Microsoft System Recovery Tool (SRT), DPM to DPM replication, and host-based virtual server backup functionality. In the case of host-based virtual server backup, a single enterprise ML on the host is required for performing virtual hard disk (.VHD) backups of any guest OSEs running on that host; however this does not include granular recovery of files or applications in the virtual machines and an individual ML is required if a DPM agent is installed on a guest to support granular application or file backups.

The most cost-effective way to license multiple System Center products is generally through the purchase of a System Center server management suite licence:

It’s important to note that SCVMM 2007 is only available as part of the enterprise suite and cannot be purchased as a standalone product; however there is a standalone workgroup edition that is limited to management of 5 physical host servers per management server console.

System Center Essentials 2007 replaces Operations Manager 2005 Workgroup Edition and is designed for management of mid-sized organisations, with some limitations to restrict it to a single installation per domain, managing up to 500 client OSEs and 30 servers OSEs. Licensing follows the same rules as for the full SCOM 2007 product – i.e. that System Center Essentials is available with or without SQL Server standard edition, that MLs are required for each managed OSE and that SQL Server CALs are not required if per-processor or inclusive SQL Server licensing is in force but are required if SQL is licensed on a client/server basis. Third party solutions can be managed and do not count towards the limits but do require an appropriate ML. A ML is not required for the OSE that is running System Center Essentials.

System Center Mobile Device Manager (SCMDM) 2008 is licensed with the standard server license plus CALs model, with both per-user and per-device CALs available. SCMDM is available with or without SQL Server licensing included and is subject to the same rules as the other System Center products that are sold with SQL Server.

The last member of the System Center family is System Center Capacity Planner (SCCP) 2007.  This is actually a free download, with capacity planning models currently available for Exchange Server 2007, SCOM 2007, Windows SharePoint Services (WSS) and Office SharePoint Server 2007.

In the next post in this series, I’ll explain how licensing works for Microsoft software running in a virtualised environment.

More new product names under the System Center brand

Two more Microsoft products in the management space got a new System Center branding last week. Whilst most of the System Center products have been new offerings, Microsoft announced that Microsoft Operations Manager v3 will be called System Center Operations Manager 2007 and Microsoft Systems Management Server v4 will become System Center Configuration Manager 2007 (these days, the acronym SMS is more often thought of in relation to the short message service in mobile telecommunications).

Microsoft management technologies – product roadmap

My recent post on Microsoft’s dynamic systems initiative (DSI) outlined the various waves of new products which Microsoft is releasing in the management space over the next few years. What follows is a summary of some of the other product roadmap information that I picked up from last Friday’s Best of the Microsoft Management Summit 2005 event:

System Center is Microsoft’s overarching brand for integration of it’s management products, in the same way that Computer Associates (CA) has Unicenter, Hewlett-Packard (HP) has OpenView and IBM has Tivoli.

Microsoft System Center Data Protection Manager 2006 is the first “System Center” branded product – launched last week in New York with an EMEA launch slated for 12 October 2005. The first release provides server backup and recovery for Windows – v2 (as part of the second wave of System Center products) will add support for Exchange Server, SQL Server and SharePoint.

Established products like Microsoft Systems Management Server 2003 (SMS) and Microsoft Operations Manager 2005 (MOM) are also part of the System Center suite and the launch of the SMS 2003 inventory tool for Microsoft updates integrates the Windows Software Update Services (WSUS) scanner into SMS – effectively a locally hosted version of Microsoft Update.

Windows Server 2003 Release 2 (R2) is due for release later this year and will bring a number of new features to Windows Server 2003:

  • New storage and management capabilities (Simple SAN, virtual disk service v1.1, common log file system, WS-Management, Microsoft Management Console v3.0).
  • Enhancements to Active Directory (AD) (federated services, ADAM in-the-box, AD as a NIS master).
  • .NET Framework enhancements (simplified data access and remoting, advanced transactions, ASP.NET v2.0).
  • Services for Unix (Unix application subsystem and utilities – no longer a separate download, database connectivity).

Microsoft are positioning R2 as a minor release – i.e. it has no kernel changes and will actually ship on two CDs, the first is effectively Windows Server 2003 with SP1 and the second has the extra functionality.

Microsoft Virtual Server 2005 R2 (formerly planned as Virtual Server 2005 service pack 1) is Microsoft’s answer for production virtual environments and will include:

  • Non-Windows guest support.
  • Network installation of guest operating systems.
  • Clustering support.
  • Greater scalability.
  • 64-bit host support.
  • Performance enhancements.
  • MOM management pack.
  • PXE booting.
  • A licensing program for the virtual hard disk (.VHD) file format.

Microsoft System Center Reporting Manager 2005 is due early in 2006 (so I guess the name will change) but is currently expected to include:

  • Integration of data from MOM, SMS and AD.
  • An extensible schema.
  • Facilitation of better business decision making.
  • Offline data warehouse.
  • Consolidated view of a multi-site hierarchy.
  • Streamlined querying.
  • Consolidated management.

Another new System Center product is Microsoft System Center Capacity Manager, a sizing solution (initially for Exchange Server 2003 and MOM 2005) which will provide:

  • Assessment of architecture choices for future deployment.
  • “What-if?” analysis.
  • Performance modelling for current deployments.
  • Identification of future bottlenecks.
  • Prediction of the user experience.
  • Understanding of the impact of changes.
  • Optimised upgrade path.

Further out on the development path are new versions of MOM and SMS. MOM v3 is expected to go into limited beta testing at the end of this year with a public beta early in 2006. SMS v4 is further out in the plan, expected in the first half of 2007 (as part of the Longhorn Server wave) with a limited beta in early 2006 which will be expanded later in the year.

Microsoft’s view is that every vendor’s management product has its agent(s), communications protocol, database and user interface, but MOM’s strength is in its knowledge, with management packs built by the product groups. Their goal is to capitalise on that strength and it is expected that MOM v3 will offer:

  • Model-based operations (more than just today’s management packs).
  • Service-oriented monitoring (using SDM models defined in Visual Studio 2005).
  • Improved task and command support.
  • Extensive software development kit (SDK) and authoring tools (making it easier to produce management packs and import knowledge, e.g. from the Internet).
  • Deep platform integration.
  • Role-based user interface.
  • Probable-cause analysis (a vehicle for managing uptime).

SMS v4 is about building on SMS 2003 (which some might consider to be the first solid SMS release), providing:

  • Model-based operations.
  • Desired configuration management.
  • IT policies and industry compliance.
  • Security interface for both intranet and Internet deployment (i.e. RPC over HTTPS).
  • Integration with Windows network access protection (NAP) to implement quarantine for patching etc.
  • Simple, role-based user interface.
  • Unified operating system deployment, pulling together RIS, ADS and the SMS operating system deployment feature pack.

Of course, much of this is still some way off, and product feature sets are always subject to change, but Microsoft is certainly making moves towards becoming a significant player in the enterprise management space – or at least for the management of their own platform.

Microsoft’s Dynamic Systems Initiative

The Microsoft Management Summit is one of Microsoft’s annual conferences and last Friday, the most popular presentations were re-run in the UK. Microsoft clearly took the event seriously, bringing across from Redmond the Corporate VP for Enterprise Management (Kirrill Tatarinov); the Systems Management Server and Operations Manager Program Managers (Bill Anderson and Vlad Joanavic); and a Director of Product Management for Enterprise Management (Michael Emanuel).

Largely due to the quality of the speakers, the event was well worth attending – particular Michael Emanuel’s Dynamic Systems Initiative (DSI) presentation. I’ve seen DSI presentations before, but this was inspirational – largely due to the charismatic way in which he described the differences between desired and actual states as “ought-ness” and “is-ness” (with associated “was-ness”, “could-ness”, “good-ness” and “should-ness”).

I’ll try to explain it all below (with a few additions from previous DSI presentations)!

It is generally regarded that infrastructure costs fall rapidly whilst performance rises (a derivative of Moore’s Law). What is less well known is that as the infrastructure costs drop, the support costs associated with supporting systems rise. Typically, 70% of an organisation’s IT budget is spent on maintenance, with just 30% on new systems. The trouble is that our increasingly well connected, but highly distributed IT systems are becoming incredibly complex. Add to that, the organisational complexity with infrastructure architects, developers, systems administrators, service architects, business stakeholders, testers, IT management and even outsourced/offshore partners – wouldn’t it be great to do something to control the management costs and let them track the decreasing cost of the infrastructure?

IT complexity and cost

Businesses tend to be dynamic. All too often, IT is not. Microsoft’s answer is the DSI, which is about helping IT organisations to capture and use knowledge to design more manageable systems and automate ongoing operations, resulting in reduced costs and more time for IT to focus on what is most important to the business.

It sounds logical enough… so why don’t we do this already? Basically because IT infrastructure architects and IT operations managers don’t tend to talk the same language! In general, designers think about scalability, security and identity but gloss over the management element. With 80% of the cost of a project committed by design decisions at the end of the design phase (but only 8% of the cost incurred), it is all too often too late to change things when they reach production and don’t fit well within an operational model. DSI is about encouraging a full lifecycle view so that operational awareness can be built into applications and services right from the initial design, using models to capture knowledge (i.e. bottling what is known for re-use) throughout the lifecycle.

The key is that systems should be designed for operations with manageability architected into the system from the outset. To do this, there are two fundamental building blocks required:

  • A generic way in which to model knowledge – the systems definition model (SDM).
  • A generic way in which to communicate with a system – WS-Management.

The SDM is basically a manifest which provides a single source of information on a system, describing:

  • What “it” is.
  • What “it” is capable of doing.
  • What “it” needs to achieve these capabilities.

WS-Management is a web services implementation of Web Based Enterprise Management (WBEM), developed as part of the Web Services Interoperability Organization’s WS-* architecture as a joint effort by AMD, BMC Software, Dell, Intel, Microsoft, Sun Microsystems and WBEM Solutions, and the first Windows implementation (WS-Management is heterogeneous) will be made available later this year as part of Windows Server 2003 Release 2 (R2).

Meanwhile, Microsoft is slowly moving the existing models within its management products over to SDM in support of the DSI and sees Visual Studio as a tool for defining the holistic structure of the application, services and system – considering management at design time to integrate service requirements during development.

By combining the application designer’s feature/functionality view of the world with the IT Operations Manager’s data centre policies and constraints, SDM models can be defined and fed through a validation process to identify errors; but a development environment in itself if not enough. Knowledge is the key to management and the diagram below shows a desired state (models, constraints, policy, prescriptive guidance, SLAs, patches) being replicated down (Emanuel refers to this as “ought-ness”) and an actual state (inventory, metrics, events, alerts, compliance, service level, results – the “is-ness”) being replicated up. The art of management is resolving conflicts between the “ought-ness” and the “is-ness” states. Furthermore, this management is not performed using an expensive tool but is actually the knowledge held by administrators and operators which needs to be re-used. The DSI vision is self-managing systems so that every application is delivered with a model which can be deployed across every Windows system.

Managing systems

SDM models are held in a models database and applied through each of the Microsoft operations framework (MOF)/IT infrastructure library (ITIL) workflows to synchronise with reality. Operational systems feed this information into a data warehouse which stores a point in time view of this reality (the “was-ness”). Taking this a step further, by applying “what-if scenarios” (“could-ness”) to this historic state, the potential (“good-ness”) of what should be (“should-ness”, or future “ought-ness”) can be modelled (i.e. capacity planning).

Of course, Microsoft is a product and technology company and so they have products which map on to this approach. Looking at the MOF model, each quadrant has associated products:

  • Changing: Microsoft Systems Management Server.
  • Operating: Microsoft Operations Manager; Microsoft System Center Data Protection Manager.
  • Supporting: Microsoft Visual Studio 2005 Team System; Microsoft Business Solutions CRM.
  • Optimising: Microsoft System Center Capacity Manager; Microsoft System Center Reporting Manager.

To summarise, DSI consists of a number of core technical principles:

  • Software platforms and tools that enable knowledge of an IT system (architectural intent; operational environment; IT policies; resource needs; across platforms)…
  • …to be captured in software models (MOM management packs; software update manifests; SDMs)…
  • …that can be created, modified and operated upon across the IT lifecycle (develop, operate, analyse/act).

In terms of product, Microsoft has currently defined three waves of products to support the move to dynamic systems:

  • System Center Wave 1 is happening now and consists of:
    • Microsoft System Center Capacity Manager 2006 (codenamed Indy).
    • Microsoft System Center Reporting Manager 2005.
    • Microsoft Systems Management Server 2003 (service pack 1).
    • Microsoft System Center Data Protection Manager 2006.
    • Microsoft Operations Manager 2005.
    • Microsoft Visual Studio 2005.
    • Microsoft Windows Server 2003 R2 WS-Management.
  • System Center Wave 2 should happen around 2006-2007 and includes:
    • Windows Server (codenamed Longhorn).
    • Microsoft System Center Capacity Manager v2.
    • Microsoft Operations Manager v3.
    • Microsoft System Management Server v4.
    • Microsoft System Center Reporting Manager v2.
  • System Center Wave 3 is due around 2008-2009, and is when the various strands of the DSI can finally be pulled together.