Application Readiness

What is Server App-V?

By: Chris Maher

Server App-V is an application virtualization technology designed for server-based applications. It enables administrators to create an "image" of an application that can be moved or copied from server to server to quickly deploy server applications. The goal is to enhance the use of cloud computing by being able to quickly spin up new server-side resources to meet growing demand.

There are some fundamental differences between Server App-V and App-V. The biggest difference is that application files such as .exe and .dll that are included in a Server App-V package are available to all processes running on the server. Objects such as COM, DCOM, COM+, WMI Providers and NT Services are also exposed to the local system. In App-V, these files and objects are restricted to the virtual application. As such, in Server App-V you must be cognizant of potential application conflicts, but communication with the virtual application by the local system is enabled.

Server App-V provides several PowerShell cmdlets that can be leveraged to automate the process of moving or copying applications to new servers. These cmdlets enable the movement of the application and the application state. The application state includes items such as application configuration and data collected while the application is running. Basically, any sort of change to the application or its environment is considered the application state. The ability to capture the existing state and move it is one of the key features of Server App-V as it enables the application to function and look exactly the same when moved to another server.

So, what types of applications are good candidates for Server App-V? Applications that leverage IIS are one category. This includes applications that install web sites or virtual directories, but not applications that make changes to IIS at the server level, as this would impact all applications on the server. Applications that use Windows Services can also be virtualized with Server App-V. Again, be careful to ensure that these applications won't conflict with other applications with the same services. Applications that use WMI are also good candidates.

In summary, Server App-V is a great mechanism for enhancing cloud computing. Using the PowerShell cmdlets, server-side applications can be moved and/or copied from one server to another seamlessly, without losing the current application state. This enables environments to be more flexible in terms of scaling to meet demand. It also allows for tasks such as routine maintenance to occur without any significant downtime.

Application Readiness for the Microsoft Private Cloud

By: Ravi Mazumdar

In my last post, I wrote about the private cloud, its definition and the importance of application readiness in migrating applications to that platform. For now we will focus on Microsoft's IaaS based private cloud technology. Built on Windows Server Hyper-V and System Center 2012 that Microsoft released earlier this year, it includes these key components:

1. Windows Server 2008 R2 with Hyper-V: Provides the hypervisor required to the host virtual machines.
2. System Center 2012 Virtual Machine Manager (SCVMM): The core management piece of the private cloud, System Center 2012 Virtual Machine Manager includes the deployment and management framework for virtual machines and other resources on the hypervisor.
3. Server App-V: Provides the application virtualization engine for server applications and allows server applications to run inside a bubble in the private cloud.

The key to successful deployment and management of a server application to the private cloud involves the following:

1. Identifying and rationalizing the application landscape: It is important to understand the application landscape to be virtualized whether they are desktop applications or server applications. In both cases there may be instances where an application has multiple versions. There is little point in spending time in virtualizing version 1.0 when version 5.0 is available. Use of advanced software asset management tools will help here.
2. Assessing the application for Server App-V compatibility: Many 3^rd party server applications are provided to enterprises as installable products in the form of MSI files that can be scanned and compatibility assessment done against Server App-V rules and guidelines. The process to manually do this is extremely complex and error prone, therefore the use of automated tools is highly recommended.
3. Converting the suitable applications to the Server App-V format: Similar to desktop App-V, Microsoft provides a tool called the Server App-V Sequencer to package the installable package (MSI) into a Server App-V package. This tool is typically run on a virtual machine running on SCVMM where it monitors the installation to capture the information and convert it into a Server App-V package. While this is a great help in the application virtualization process, streamlining the steps between assessment and conversion and then automating some of them will increase efficiency.
4. Deploying virtualized application to SCVMM: In order for the application to be available to users, it must first be published as a virtualized package to SCVMM. While the sequencer does not allow you to do this, SCVMM provides a number of tools in the form of PowerShell scripts as well as the SCVMM UI that allows you to import a package into its library, create a service template, and deploy it as a service. Again, the process can be made more efficient using tools that provide a simple user interface and automate part of the process.

In conclusion, it is clear that private clouds are here to stay and that they present many benefits as IT transforms the way data centers are managed. Adopting the new wave and taking advantage of the benefits will put you ahead of the curve.

Application Readiness in the World of Private Clouds

By: Ravi Mazumdar

First of all, let us try and understand what a private cloud is. A common misperception appears to be that by simply adding virtualization and converting physical boxes to virtual machines will make it a cloud instance. Clearly, there is more to cloud computing than adding a few hypervisors to replace existing physical servers running on a rack.

Gartner's official definition of cloud computing is that it is "a style of computing where scalable and elastic IT-enabled capabilities are delivered as a service to customers using Internet technologies." Furthermore, it defines a private cloud as "a form of cloud computing where service access is limited or the customer has some control/ownership of the service implementation." This is where you deploy cloud-based solutions within your own network and data center with access limited to or controlled by your IT resources. It is estimated that 45% of all cloud deployments are based on private cloud platforms.

There are numerous vendors, including Microsoft and VMWare, delivering multiple types of private cloud solutions including those that are based on an IaaS type of architecture as well as PaaS based architecture with IaaS being the more common and chosen architecture. It presents the following benefits:

1. Pooled Resources: In a private cloud, the key resources—compute, storage, and networking—are pooled and abstracted into units that enable you to dynamically provision and scale applications and resources. This will prevent underutilization of infrastructure distributed in siloes across the enterprise and better return on the hardware investment.
2. Elastic: Resources can quickly be expanded or contracted through automation or workflow, so IT services can be scaled up or down almost instantly to meet business requirements.
3. Self-Service: Your customers can request, configure, and manage IT services through interactive portals that allows for automated provisioning streamlining the application deployment and update process.
4. Usage Based: With resources as services, usage can be metered so that you and your customers pay for only the resources that are actually consumed.

While the above benefits are great, bottom line is that your business needs your applications to run in this environment with maximum uptime i.e. it's really "All about the App". When we refer to apps here we mean server applications typically running in your datacenter catering to various needs of your enterprise users, e.g., CRM systems, accounting software, corporate expense tracking system, etc. Some of these are obtained from 3^rd parties while others may be developed within the company.

In both cases, best practices recommend the adoption of the application readiness process where application compatibility assessment is done on the targeted application, migrated to the required format, and published to the private cloud. These ensure smooth transition of the apps to the private cloud.

An Increasingly Hybrid App World

By: Steve Schmidt

Attending a session on Application Management at Microsoft TechEd in Orlando, FL, it strikes me once again that despite a general desire to simplify computing environments, IT organizations find themselves in a world with a more, rather than less, diverse set of applications. The variability exists on many dimensions:

• Application types – These include traditional on-premises apps as well as SaaS apps, and those apps that will exist as a combination of software + service.
• Application packaging types – Applications exist in multiple formats including MSI, application virtualization (including client App-V and Server App-V), and virtual appliances. In addition, an increase in adoption of user profile virtualization is creating another set of templates and data sets that accompany applications.
• Delivery methods – Apps can be installed locally, streamed as needed, and/or hosted in a datacenter (on-premises or off-premises). On another dimension, apps can be delivered as needed as a "layer" on top of an OS image.
• Target platforms – The app can be run on a traditional client, mobile device, or server environment. Many variants of each exist.
• Application counts – The number of applications in use seem to be increasing, based on the requirements for specialized functionality and the fact that additional personnel are now bringing them into the organization.
• Licensing models – The traditional models of per user, per server/CPU, concurrent user, and others have now been augmented with a range of usage-based metrics.

A key reason for this increase is that we are in a state of transition, fueled by multiple major market trends. Virtualization and cloud computing are yielding applications that appear in new formats, such as the virtual appliances and virtual applications. They also give rise to new or redesigned apps built explicitly for multi-tenant hosted delivery. Concurrently, the mobility trend is resulting in many new device types in the enterprise environment, running different apps or apps tailored for those hardware/OS combinations. Additionally, mobile devices bring some different connectivity capabilities and usage patterns which frequently affect how applications will be delivered or presented to them. A third significant trend is the consumerization of IT. End users are often given the ability to select applications they need, resulting in the influx of new applications in new formats. Coming from different sources, there may be other nuanced differences in the way they were versioned, prepared or delivered. The fact that new trends are adding variability is typical, especially in early stages of the trend adoption. It is also likely that the life of many of the incumbent technologies will have a "long tail". Full displacement of key technologies is rare; there are still many uses for mainframe, midrange, and first generation client/server solutions, even as each generation of successive technology was introduced.

So, what should IT organizations do? Some of the planning implications are:

• Application assessment - With so many application packaging format options to consider, a mechanism to determine format suitability for particular applications and users should be considered. These include both application virtualization formats and virtual hosted desktop environments. This compatibility testing should be automated so it can be applied to a broad section of the application estate quickly.
• Flexible processes and tools – Processes and tools should be dynamic enough to create and manage multiple formats. Integration with many other systems, including configuration management and IT Asset Management systems, will be important. Those other systems will be sources of complementary data, as well as delivery vehicles for the apps, whether in traditional client or hosted approaches. Since some organizations will continue to test and use different application packaging options (by department, by region, etc.), tools and processes should be neutral in their support of new technologies from different vendors, supporting all the key options available.
• Coordination – The effort should include centralized management in order to avoid duplication of effort across departments and to retain some control within the IT organization, at a minimum. The functionality in this area can include workflow management for handoff between key process steps and the logging of results at each step, a central repository for the applications and their relevant metadata, and reports to demonstrate progress and future plans.

A hybrid application world is here, and it is increasing in scope. That isn't necessarily bad, as long as we recognize it and prepare for it.

Symantec Vision 2012 – The Information-Driven World is Rapidly Changing

by Toby Martin

Having just returned from the Symantec Vision Show in Las Vegas, I wanted to share some commentary that I gathered while at the event. As the threat landscape evolves and end-users are rapidly adopting new technologies, Vision 2012 provided some insights about where things are headed:

 

• The partnership between Symantec and VMware was spoken of often, and fondly, by the Symantec team, their partners, and end users who are intrigued to see where this is going. Reaching into the Hypervisor from within the Symantec Suite definitely opens up many possibilities and should make this very interesting to watch as it evolves.

 

• According to Forrester, expect 200 Million mobile devices in corporate environments by 2016, and the growth will be even faster in the amount of data housed. Cross-platform support has won over most of the major players and is especially strong for iOS. Several apps and versions coming this year are supporting iOS first, before Android or Windows mobile. Of the multiple demos I got to see − mostly on Apple devices − I found the separation of application access from data the most interesting. This is being done with forced authentication and inter-application security policies blocking sharing of data amongst un-secured devices. 

 

•  Another common theme was persistent vs. non-persistent environments, and all that they entail. It appears that demand for persistent environments is increasing as a result of Bring Your Own Device (BYOD). The Generation X & Y users don’t even look for security policies, they just bring in their devices and hook them up, so you’d best secure the data because it’s already too late to secure the devices.

 

•  The ‘Cloud’ was pervasive with many thought leadership and cloud sessions positing that in 2-3 years there won’t be sessions on the cloud, but instead it will be everywhere in nearly all the sessions. The mindset is changing from ‘security and compliance tools FROM the cloud, to security and compliance tools FOR the cloud.’ Much of the discussion about the cloud centered on the emerging threat landscape and the fact that attacks done with applications are not only targeting enterprise or government. As a result, endpoint protection on virtual desktops is increasing in importance to abate the ever-growing risk.

 

•  Symantec recognized that they have several portals and they committed to working on unifying them, which is good news for users. While they do provide a means of interacting through APIs and the like, it is still incumbent upon the end user to get this to work. Some developments are:

-   Nukona acquisition (2Q11) will be moving in the direction of single-sign on for application access

-   Regardless of the device on which the user logs in, persistent security settings will follow, thereby achieving the dream of managing BYOD from anywhere, anytime

 

So to sum up, the pervasiveness of the cloud and associated services required to secure delivery via the cloud and via mobile devices was top of mind and topic and is the direction we will see Symantec continue to invest.

Observations from the Microsoft Management Summit (MMS)

by Toby Martin

It was a great week with some excellent sessions; here are some of my observations:

• Bring Your Own Device (BYOD) is creating more security related work for the IT workforce
  • Good idea on the difference between governed and managed devices, quadrant resonates with everyone because you cannot manage mobile devices in the traditional manner
  • New roles, scopes and permissions are becoming more granular as the mobile workforce and IT collide over securing data
• Still major worry about licensing for mobile and cloud delivery
  • Microsoft will be simplifying their license models, per several sessions and discussions, which is good news for complex user environments
  • Licensing in the cloud is a big gap, and it’s yet to be figured out
• System Center Orchestrator has made leaps and bounds forward
  • Run book automation (RBA) discussions were overheard several times, as was the idea of workflow templates
  • In addition to the app model, System Center 2012 Orchestrator was featured prominently in several sessions and many discussions on the floor and in the hallways
  • The need to plug into workflow models will drive new innovations in connectivity with platforms
• Start thinking of Microsoft as a platform and your survival will depend on your innovations
  • Take more control of your own destiny by using the tools they provide to leverage the System Center platform
• Flexible worker and lifestyles have huge requirements in new infrastructure and management
  • System Center 2012 Configuration Manager (CM) deploying to iOS is a pretty amazing recognition of reality – Microsoft is now less anti-iOS than previously thought (despite having their own competing platform)
  • Bold move by Microsoft to remove BYOD issue
• Cloud reality sinking in – the hybrid will be here for a long, long time
  • Progress in Azure, Intune, and other cloud platforms all make it clear that Microsoft is  investing massively there – and correctly
• Structure of cloud not nearly as important as the delivery of services

• System Center Virtual Machine Manager (SCVMM) will become a big continue to erode the leadership position of other virtualization vendors
  • They’re advancing the server management capabilities at a rapid pace

• Win8 was not as much a focal point – Microsoft spent more time talking about business objectives to be solved rather than the O/S

• This is going to be a busy year for keeping up with all of the releases coming up from Microsoft

Next year – New Orleans, Laissez les bons temps rouler in June!

6 Key Insights About New Disruptive Technology

by Randy Littleson

Just read an interesting article entitled “How are CIOs looking at today’s disruptive tech trends?” that summarizes a recent roundtable on digital strategies featuring the CIOs from American Express, Bechtel, Chevron, Eastman Chemical, Eaton Corporation, the Hilti Group, Holcim, Nestle, Sysco, and Time Warner Cable. 

It’s an interesting look into the challenges these CIOs face as a result of the major disruptive technologies that are getting so much press these days (mobility, social media, consumerization, cloud and big data).  The article details 6 key insights about new disruptive technology that I think are insightful for anyone in IT (there’s a link to a 17 page report within the article as well).

One of the six insights really caught my eye: “Designing for loss of control” is one of IT’s key challenges.  In essence, it acknowledges that IT can’t keep up, but is still going to be held accountable for security, reliability and performance.  I would add spend management to this as well.  This is very true and matches what we see every day around applications.

IT is being asked to deliver applications faster to their business users, cope with rapidly changing access requirements (e.g., the same app accessible on a laptop and an iPad) while trying to leverage new technologies (e.g., cloud, virtualization) to meet these needs and save money.  Most organizations don’t have mature application readiness processes because in a more static world, it just wasn’t the priority.  But, in today’s rapidly changing, heterogeneous and increasingly complex environment, they need to establish a robust process to automate and manage application migrations and on-going operations to accelerate Windows application availability for on-premises, virtual or cloud deployments.  This process needs to automate and manage compatibility assessment, application packaging and deployment, which now excludes the expectation of applications being virtualized, deployed in the cloud and accessible to employees via a self-service enterprise app store.

In addition, given the strategic importance and aggregate spend of applications, IT must establish processes for software asset management and license optimization to ensure they are buying what they need and using what they bought.  But this too is becoming increasingly difficult with the influx of new technologies such as virtualization and bursting to the cloud which makes tracking and understanding application usage that much more challenging.  Processes need to be established to continuously ensure usage is in compliance with software license agreements and to provide the organization the insights they need to manage how many software licenses are required based on actual usage on a global basis.

IT will continue to be held accountable for ensuring software license compliance and driving increased application usage despite the disruptions that all of this external technology change represents.  Establishing business processes to manage application readiness, software asset management and license optimization can go a long way to helping IT “design for loss of control” by ensuring that continuous processes are in place to strategically manage application usage.