Microsoft System Center 2012: Application Management

In this series of looking at Microsoft System Center 2012 (SC12), we at Uncommon want to give you a solid overview of SC12 without getting too far into the technical aspects of how to implement or work in the platform.  As a review, let’s look at what programs make up the System Center 2012 platform:


  • System Center Operations Manager
  • System Center Configuration Manager
  • System Center Endpoint Protection Manager (now integrated into Configuration Manager)
  • System Center Virtual Machine Manager
  • System Center Data Protection Manager
  • System Center Orchestrator
  • System Center App Controller
  • System Center Service Manager
  • System Center Advisor


For this month, we will be looking at the Application Management portion of SC12.  Of the programs listed above, the Application Management is made up of three: Virtual Machine Manager, App Controller, and Operations Manager.


(Next month we will be looking at the Service Delivery & Automation section: Orchestrator and Service Manager.)


Before we dive into the highlights, it will help the reader to understand the primary driving philosophy behind Microsoft’s new vision.  All of Microsoft’s recent moves revolves around what they call “IT as a Service.”  The following key points from Microsoft’s TechNet blogs define this new catchphrase:


  • If you’re referring to Infrastructure as a Service (IaaS), you're thinking about your datacenter as a set of pooled resources (including compute, network, and storage), not in terms of individual hosts or VMs.


  • If you’re referring to Platform as a Service (PaaS), you're talking about building applications that will then be delivered as a service. The platform provides all the required building blocks for your app. (i.e.: Windows Azure)


  • Between IaaS and PaaS, we believe that there is a new category emerging in the marketplace called IT as a Service, a way that will transform how customers consume IT and will deliver a completely new cost structure at a much higher level of business responsiveness.


Microsoft believes that the emerging hybrid environments will become the norm for IT.  The common theme in these hybrid environments will be integrated physical, virtual, IaaS, and PaaS management that will the consumer optimize the ROI on the IT budget.



Microsoft’s definition of the Virtual Machine Manager (VMM) is quite straightforward, but a little simplistic in actually telling you what it does:


  • System Center Virtual Machine Manager (VMM) helps enable centralized management of your physical and virtual infrastructure, increased server utilization, and dynamic resource optimization across multiple virtualization platforms.


Their goal for SC12 is to produce cost savings both on capital equipment and on managing the virtualized environment.  Part of these savings will come from automated, orchestrated updates to all machines in a cluster.  For example… when an underlying Hyper-V cluster is receiving a service pack update, any services running will not have to be turned off allowing the update to take place while the service is still available.


Another couple of examples of Microsoft’s drive to uninterrupted services are Live Migration and Live Cloning.  Live Migration allows a virtual machine to be transferred from one host to another without any downtime.  Live Cloning is similar in that it allows exporting a virtual machine without downtime… you can then avoid creating and configuring new virtual machines to match existing virtual machines and instead simply stand up a new VM which already matches the others in the cluster.


The common management toolset provided in SC12 empowers solutions for IT to manage their private and public cloud application and services.  These tools can deliver a self-service infrastructure allowing for simple creation of a private cloud.  This optimizing usage of your datacenter investments enables pooling and dynamically allocating your compute, network, and storage resources.


Those of you who are familiar with VMM 2008 will be happy to hear that the 2012 version still manages and deploys hosts of VMs.  However, the focus of the new 2012 is on private clouds and service templates with the goal being that an administrator (or even an end user) can deploy a service or application to a private cloud without having to be able to understand the physical layer beneath the application.


Some other updates from the previous offering are:


  • Where VMM 2008 was managing virtual machines, now VMM 2012 includes managing the physical environment as well


  • Support for heterogeneous hypervisors enables more flexibility in the data center


  • VMM 2012 creates network abstractions by taking network resources and exposing them to the end users


  • VMM 2012 enables effective cloud management by taking the underlying compute, network and storage resources, and dividing and delegating them to individual users


  • Services or applications can be deployed to end users as individual virtual machines or pools of resources in your private cloud environment


  • You can re-allocate the workloads to the different virtual machines that need the resources based on utilization rates of the services running on top of them


  • Power Optimization enables you to shut down any extra servers when they are not in use to reduce power consumption in the data center, and similarly bring the servers back when needed


To sum up Microsoft’s Virtual Machine Manager in System Center 2012, VMM allows you to manage those multiple Hyper-V computers more or less as a unit; you can apply policies to the cluster as a whole instead of one machine at a time.  You can monitor your IT cluster (private and cloud) from one pane of glass and improve your management by working with those computers as a single unit instead of individual units.



Again, let’s start with Microsoft’s definition of System Center 2012 App Controller:


  • App Controller lets you manage applications across the private cloud and the Windows Azure platform from a single console. You can manage application components in the context of the service that it represents to the business, so that you are managing services rather than servers


Where VMM was designed to work with the machines in the hybrid environment, App Controller (AC) was designed to work with the applications in that environment.  The target users for AC are not administrators in particular (although some admins may use AC for admin tasks); instead, AC users will most likely be the application or service owners.  These are the people who deploy and manage an application or service (note: not end users who use the application or service).  The App Controller allows the owners to have a self-service way to deploy new instances of a service or application without requiring them to deal with the underlying physical environment.


However, AC cannot perform in a vacuum; nor can it be used to create new objects from scratch (except for service instances).  Anything you work with in App Controller must first be prepared in VMM. System Admins must first create VM templates, guest OS profiles, hardware profiles, application profiles and so on. To deploy services through App Controller, a VMM administrator must create a service template and deployment configuration. Self-service user roles also should be created in VMM and associated with one or more private clouds and quotas.  This prep work by IT allows for the environment that AC users will work in.


Note: App Controller doesn't have its own security infrastructure.  AC relies completely on security settings in VMM, so available options for a user in App Controller depend directly on the rights and permissions that are assigned by the Sys Admin to the user in VMM. Authentication is performed by using a web-based form, but you can opt to use Windows Authentication in Microsoft IIS to achieve single sign-on (SSO).


Cloud computing is still evolving as the underlying structures become more stable and robust.  The hybrid approach continues to be the go-to IT computing model for the foreseeable future.  Microsoft designed the App Controller to enable IT a way to strategically connect both on-premises System Center private clouds with off-premises deployments in both Windows Azure and third-party cloud hosting providers.


In conclusion, Microsoft System Center 2012 R2 App Controller is designed to manage applications and services that are deployed in hybrid environments from the application/service owner's perspective. By providing a self-service experience allowing the owner to configure, deploy, and manage virtual machines (VMs) and services, AC enables higher efficiency and lower IT cost.



Microsoft defines System Center 2012’s Operations Manager as:


  • Operations Manager provides deep visibility into the health, performance, and availability of your datacenter environments – across applications, operating systems, hypervisors and hardware – through a single interface


This definition is pretty solid and really does state what Operations Manager does… it lets you answer the question, “What the heck is going on?”


Frequently, when network administrators hear of an issue with an application, they have a tendency to say they don’t have any proof that there’s a problem, and request proof before they will look into it.  Operators need insight into the network layer; and with that insight, operators need to identify a root cause quickly.  This is where network device monitoring in Operations Manager comes in. Microsoft decided to give good visibility to operators, without necessarily digging deep into the network traffic itself. You want to be able to monitor the devices and figure out what is going on with them, whether they are up or down, and what is being affected by their state.


Out of the box, Operations Manager has deep monitoring of over 2,000 devices and can monitor these devices to the level of memory utilization, processors, network traffic, and what is connected to the device. All of this information provides indicators of what’s happening when one of the network devices goes down. Now you can see how one device going down can affect other devices, as well as the servers that are within the system.


In total, the network device monitoring in Operations Manager supports bridges, firewalls, load balancers, routers, and hosts.  You can discover all of these devices and monitor all the activity from server to switch, and from switch to switch.  This information allows you to build a topology map so you can get a mesh where you can see how everything is stitched together, and you can do several network hops and see how all this looks in the overall environment.  Operations Manager can also discover VLAN memberships, HSRP groups, and server NICs, in addition to monitoring the ports, interfaces, and memory for the individual devices. Overall, this gives you a holistic view of all the devices that are in the network.


Microsoft has done a lot of work on dashboards to make sure you can visualize this information in a very easy and consumable way.  As well, you can look at response time, and monitor all the individual devices, and all the ports on those devices, and see what the response time for those ports is. These metrics provide an indication of what is happening in terms of the load that is being placed on the individual device, as well as the timeline when the load is occurring.


Those dashboards are made up of many individual functions.  Here is a bullet list of some of the more interesting ones:


  • Fabric Monitoring
    • A close integration between System Center 2012 R2 Virtual Machine Manager and System Center 2012 R2 Operations Manager introduces System Center cloud monitoring of virtual layers for private cloud environments.


  • Fabric Health Dashboard – Monitoring the Health of Private Clouds
    • The Fabric Health Dashboard shows a detailed overview of the health of your private clouds and the fabric that services those clouds. The dashboard helps you answer questions like “What is the health of my clouds and the fabric serving those clouds?”


  • Network Node State
    • Displays the health state of network nodes (devices) that are relevant for the cloud you selected.  The Active Alerts and Number of VMs fields on this dashboard help indicate which issues are having the greatest impact on your cloud and can help you prioritize your work.


  • Fabric Monitoring Diagram View
    • Displays health states of cloud and on premise environments.  The Diagram view gives you a diagram of the entire infrastructure and shows the health state of each part of the fabric. The Diagram view helps you answer questions, such as “What is the health of my entire fabric?”


  • Microsoft Monitoring Agent
    • Microsoft Monitoring Agent collects traces on demand or can be left running, which monitors applications and collects traces continuously.  Microsoft Monitoring Agent can be used together with Operations Manager or can be used as a standalone tool for monitoring web applications written with Microsoft .NET Framework.


  • Support for IPv6
    • The Operations Manager console can take IPv6 addresses as input for Network Discovery and display IPv6 addresses in the network-related views.


  • Java Application Performance Monitoring
    • Java Application Performance Monitoring lets you monitor Java application performance and exception events. You can set method and resource timing for performance events, stack traces for exception events, and set Java specific counters for events. Additionally, you get Operations Manager level alerting on Java application server counters.


  • System Center Advisor
    • Advisor collects data from your installations, analyzes it, and generates alerts that identify potential issues (such as missing security patches) or deviations from identified best practices with regard to configuration and usage. Advisor also provides both current and historical views of the configuration of servers in your environment. Ultimately, Advisor recommendations help you proactively avoid configuration problems, reduce downtime, improve performance, and resolve issues faster.


  • UNIX and Linux Monitoring
    • UNIX and Linux agents for Operations Manager are now based on the Open Management Infrastructure (OMI) open-source CIM Object Manager.  Debian GNU/Linux 7 is now supported by the Universal Linux agents and Management Packs.


Yes… you read that last bullet correctly.  Microsoft System Center 2012 does support open source environments.  This should really come as no surprise to those who have been checking up on Windows Azure… open source OS are now available natively in Azure.