Lately I've been thinking about the pricing models of Cloud Computing Vendors and wondering when they'll move to a more granular pricing structure like the mobile phone market.

Much of the hype around Cloud Computing is around paying for what you use including compute time. That's a pretty powerful message, but what does that really come down to? While Amazon Web Services (AWS) focuses on Infrastructure as a Service (IaaS)/IT buyers and Microsoft Azure focuses on Platform as a Service (PaaS)/Developers, they both end up delivering the same basic service: Virtual Machines on an Hourly Basis. Yes, there's more to it than just that, but let's focus on the compute part shall we.

As the developer delivering my application via Azure or an IT person managing an application delivered on AWS, I must define how much capacity I need and ensure I turn on the right amount of Virtual Machines (VM). While there are management tools and API's to automate this process, it still comes down to turning on and off VM's. By doing so, it impacts my overall bill (so does data usage, data traffic, and more…but again, let's stay focused on CPU/Compute) and it's charged on an hourly basis.

The only On-Demand part of this is that I turn it on when I want, but that doesn't actually mean that I need it now. It could be hours until a person on the Internet wants to use my application or my server deployment needs to react to a request (e.g. incoming email), yet I still pay for that time the VM is up and running.

The future of On-Demand computing, pay for compute, and cloud computing might still very well be VM based, but done in a way where you only get charged for actual usage, regardless of the fact that your application is dormant for hours, days, or months. When a request to your application is made, it will be ready and your bill will reflect actual CPU cycle usage. Perhaps down to the minutes and seconds. Almost like your mobile phone bill.

The iPhone phenomenon is driven by many innovations made by Apple and now many software makers are scrambling to see how they can replicate that success for their own application and market. It's more than just focusing on the mobile market and the iPhone specifically, the same model and success could be found in other software and service delivery models. That said, I think many software makers are making a mistake by assuming the AppStore's success is because customers were looking for more applications and willing to pay for it. I'm sure that's what all software vendors would like to think, but this isn't really listening to what customers where frustrated with previous mobile devices.

"Back in the day" when there wasn't an iPhone snuggly tucked into your pocket or blazing a custom ring-tone in your purse, we were carrying a Blackberry, Palm, Nokia, or Windows Mobile's. Each of these devices had the ability to have custom applications deployed on your device. But this of course required a cable, custom software deployed on your PC, and even worse may require additional third-party software to make your game, business applications, or utility to work. Sure you had a slight issue of where to find these applications, purchasing them was a hassle at best, and then the applications weren't that user friendly. Taking all of this into account, it was a real pain for an average user to "add value" to their phone beyond what it came with. Users were asking for an easier way to find and deploy their application. It's this basic demand from users, both business and consumer, that the iPhone plus the AppStore had a profound impact on the market.

The market for applications on the iPhone is driven mostly by user's acceptance to deploy applications to their mobile device. Sure the Software Developers have a great device to target, lots of passionate users, and a Software Development Kit that makes building applications for the iPhone an easy business decision and thus creating a huge library of applications available. But it's the simplicity in deployment that made the difference.

My main point here is simple. Make it "easy" to deploy applications and users will be willing to deploy and use them. To take it one step further, it removes the notion of installation and deploying. Now users simply find, purchase, and use. Really, isn't that exactly what both users and software makers want?

Accelrys is looking for beta testers for its upcoming release of Pipeline Pilot. The beta is tentatively scheduled for the week of December 7. We’d like feedback on a number of capabilities. Of particular interest will be to find out if these capabilities behave as expected in a true customer setting and if behavior for existing protocol is unchanged after an upgrade. If you are interested in becoming a beta tester, and you can commit at least 4 hours of your time before the end of the year, please send an email to Ton van Daelen (tvandaelen@accelrys.com) with the following information:
• Company:
• User name:
• Capabilities interested in testing:

For more information on the Capabilities, click on the link below which will take you to the Accelrys Forum and provide more details:

http://forums.accelrys.org/eve/forums/a/tpc/f/7421036442/m/325109862

My notes from Day 2 at the Microsoft Professional Developers conference in Los Angeles on November 18, 2009.

• 8:30 - Steven Sinofsky takes the stage!
• Kicking it off talking about Windows 7
• Video time – “Why we share your pain”…Funny!
• Explaining how Microsoft uses Windows 7 Telemetry to learn more about how customers use Windows 7
• 4,753 Code changes driven by Windows Error Reporting
• 1,729,890  Send Feedback Reports
• 46, 447,784 Aero Snap & Shake features used in last 6 months
• Video time – Showing some of the usability testing done by Microsoft for Windows 7.
• 8:53 – Michael Angiulo, Planning and Ecosystem Team
• Showing last years Netbook (ASUS eepc)
• Netbook is the fastest growing PC’s
• Explaining that applications can perform better for the use by using Windows 7 specific functions.  Basically….Microsoft asking developers to play nice in Windows 7 so customers don’t think the OS is the problem.  Good luck with that.
• Showing off some cool PC hardware (media acceleration, Gaming PC’s) that take advantage of Windows 7
• Showing off the new super thin Dell laptop (didn’t catch the name) and it has many of the new sensors supported by Windows 7 (heat and light sensors)
• Showing that Google, Firefox and others are already using Windows 7 API’s like Jumplists and Flick.
• 9:08 Laptop for all attendees! 
• 9:09 Steve Sinofsky talking about Internet Explorer 9
• Advances in Standards, Privacy, and Innovations
• IE “should” be the very best browser for Windows
• IE 9 Update: HTML 5, improve JavaScript performance, Hardware Accelerated Graphics and Text
• Showing internal build of IE9 running acid3 Browser test (FAILED 32/100) & CSS Selector Test, graphics rendering using hardware acceleration  (no need for the developer to do anything…Improvements in the browser work by Microsoft will benefit the dev/user)
• 9:22 – Scott Guthrie takes the stage (Red Shirt of course!)
• Silverlight 3 shipped 3 months ago
• Showed Expressions Sketchflow
• Sunday Night Football is using Silverlight with full DVR support
• Will be used in Winter Olympics and Victory Secrets
• Siemens is using it with Healthcare applications
• 45% penetration into the browser marketplace
• Silverlight 4!!
• Focused on: Media, Business Applications, Beyond the Browser
• Media Enhancements
• Access to Webcam and Microphone (Demo built by Archetype)
• Multicast Streaming
• Output Protection
• Offline DRM
• Showing Enhanced Media support (ie. Smooth Streaming) on the iPhone!
• iPhone tab added to Publishing Point within the IIS Admin Tool
• Application Development
• Printing
• Rich Text
• Clipboard Access
• Right Click
• Mouse Wheel
• Implicit Styles
• Drag/Drop
• Bidi & RTL (Right to Left)
• HTML
• Commanding and MVVM
• Additional Controls
• Data & Networking
• Share Assemblies across SL and .NET 4
• Data Binding Improvements
• UDP Multicast Support
• REST Enhancements (with ADO.NET Services)
• WCF Improvements
• WCF RIA Services (Simplifies data transfer between client and server)
• Visual Studio 2010
• WYSIWYG Design surface
• XAML IntelliSense Improvements
• Data Binding, Layout, Styles
• WCF RIA Services Integration (including SharePoint connectivity)
• 9:50 – Scott Hanselman, Principal Program manager takes the stage for a demo
• Using Visual Studio 2010
• Creates Entities and Classes
• Leveraging WCF/RIA Services
• 9:58 – Samad Wahedi, CEO from SnapFlow to do some Demoing  (Demo built with InterKnowlogy and SoftSource)
• Built a Workflow based application that leverages Silverlight
• They tried AJAX and Adobe Flex, but chose Silverlight because of their developer knowledge and the power of the platform
• 10:02 – Scott back on stage and talking about Silverlight our of the browser
• New Silverlight 4 features focused on “Out of Browser” support
• New Windowing API’s
• Notifications Popups
• HTML Support
• Drop Target
• Trusted Application support (Run outside of the Sandbox).  Checkbox enabled by the developer and End User must accept upon install
• Custom Window Chrome
• Local File System
• Cross-Site Network
• Keyboard in Full Screen Mode
• Hardware Device Access
• COM Automation Support
• Windows Only
• Location API
• Silverlight Performance Updates
• Twice as Fast
• 30% faster Startup
• New Profiling Support
• 10 Second installation
• Will support Google Chrome
• 10:07 – Brian Goldfarb, Director Dev Platform & Tools doing a demo of Silverlight application that is “Out of Browser” that shows feeds from Facebook
• Using new Silverlight Facebook API’s which shipped last week.
• Showed device support (Took a picture, plugged it in, and the Silverlight application allowed you to connect to the device)
• Showed Multitouch support
• 10:15 – Scott back on stage
• Schedule
• Beta: NOW!  http://silverlight.net
• Release Candidate: TBA
• Final Release: H1 2010
• 10:18 – Scott’s Off … Kurt DelBene, Senior VP on stage to talk about Office and SharePoint 2010
• Blah, Blah, Blah….Corp Message, seen it, heard it, give us some Dev Updates and Code!!!
• 10:38 – Derek Burney, General Manager of SharePoint going to do some demos (Yes! Finally)
• Showing new projects available within Visual Studio 2010 for Office (SharePoint Solution Projects, WebParts, etc.)
• SharePoint code has integration with Source Control
• SharePoint can now be deployed on Windows Vista and Windows 7 to simplify the development environment while developing for SharePoint
• Don’t need to be a SharePoint administrator to deploy a SharePoint project to it since it will run in a sandbox
• Set Breakpoint and “F9” to debug and deploy the solution to a SharePoint server….Visual Studio 2010 does it all for you now!!!!
• Showing Business Connectivity Services (BCS) in SharePoint 2010 pull data from SQL Azure.  It’s an interesting idea and I’m sure it will turn into a trend within the next year or so as more data gets stored in the cloud.
• 10:56 – Kurt is back on stage
• Announcement Time – Microsoft and SAP Duet Enterprise for Microsoft SharePoint and SAP
• Consume and extend SAP through Microsoft SharePoint 2010
• Allow’s developers to create applications using SAP data through SharePoint 2010.
• Announcement Time – Outlook Social Connector (looks like a Google Wave light competitor….kinda)
• SharePoint
• Windows Live (Spaces)
• LinkedIn
• Where’s Facebook and Twitter?
• It’s an SDK, so other developers to extend into other social networks
• Derek shows a demo of Outlook Social Connector
• 11:03 – Kurt invites us to Mix 2010 in Las Vegas, March 15-17, 2010 at Mandalay Bay http://visitmix.com

Follow Me on Twitter

This is day one of the Microsoft Professional Developers Conference (PDC).  My goal is to keep posting to this single blog entry with updates from the conference based on the keynote and sessions that I attend

Keynote – Ray Ozzie

• 8:30 - Ray takes the stage to show off the high level vision for Microsoft products and services.  Nothing interesting to show off yet.
• 8:41 - Loic LeMeur (Founder & CEO) from seesmic showing his product using Silverlight.
• User Interface looks a lot like TweetDeck (which is built on Adobe Flex)
• Announced that the product (for Windows) has shipped today.
• Created as a development platform for other developers to extend.
• 8:46 - Ray back on stage discussing Windows Azure
• Windows Azure Fabric Controller is what enables the cloud in the datacenter…. It’s “like System Center”
• Said that Java, Eclipse, PHP, and mySQL will be available for use within Windows Azure
• Showing Timline for Azure
• Oct 27th, PDC 2008 – Azure is announced
• Nov 17th, PDC 2009 - Today
• Jan 1 – Production platform deployed
• Feb 1 – Begin billing
• Provide multiple sized VM’s within Azure
• Supports multiple platforms
• Java with TomCat
• php
• mySQL
• java
• Zend Framework
• Shows datacenters of Windows Azure as Pairs within a Geography
• Chicago/ San Antonio
• Dublin/Amsterdam
• Singapore/Hong Kong
• Azure Xdrives – Mountable durable storage that is an NTFS mountable device.  Not clear yet if this is directly available on a local computer, but I suspect that’s true.  Sounds like Amazon EBS.
• 8:55 – Matt Mullenweb (Founder, Automattic) from WordPress is now onstage showing off Wordpress.  What is this dude doing here?
• Showing Wordpress running on Windows Azure
• Using php, MySQL on Azure…. Very nice!
• Matt (Didn’t get his last name) from FailBlog (and other sites) discussing their services and scale issues.
• Showing 8 million page views a day network that
• 100,000 votes a day driven by users
• Launching a new site TODAY!  http://oddlyspecific.com built on Windows Azure and SQL Azure.  This makes it REALLY easy to scale
• Using a Windows Azure Storage Plugin
• 9:02 – Ray is back on stage. 
• Microsoft Pinpoint is a new Marketplace for users and IT to find new services.  Integrated into Azure Portal today and soon will be integrated with the Partner portal and on the core Microsoft sites. 
• New project called “Dallas” which is a data catalog service (Discover, Explore).  Dave Campbell to discuss further:
• Some are public and some are private data sources.
• Allows users to provide comments on the data
• User can “try” the data and even requires you to “sign” a EULA specific to that data source.
• Dallas can help you to preview and navigate the data, so you don’t have to learn the entire structure just to evaluate it.
• Can use Excel PowerPivot to explore the data.
• When pointing to the service via Visual Studio, you get an object collection built for you and allows you to easily connect and use via ADO.NET.  I’m sure there’s more going on here, but it really does make it easy to discover, explore, and utilized the service.
• Showing a WPF application pulling data via Dallas and we used some provided 3D glasses (like the ones from an old 80’s movie) to view data from the NASA Mars Rover data.  Very cool!  Be a Martian
• 9:14 – Ray back onstage
• Discussing that “This Administration” (talking about the US. Government… Obama) is interested in cloud computing and access to innovation.
• Vivek Kundra, CIO of US Government is connected via Video chatting with us.
• Explaining on how innovation is happening as public data is being made available and Cloud Computing.
• Showing how Data.gov is used to search for jobs based on location.  Showed this using an iPhone (funny).
• Asking the PDC crowd to developer applications “for the good of the republic”.
• Saying how this is the early days of cloud computing.  “Three Screens in the Cloud”
• Ray asks that we remember 3 things
• Bet on Windows 7
• Bet on Windows Azure
• Take a moment to think how Azure, Pinpoint, codename “Dallas” will impact our world.
• 9:24 – Ray is done…now a funny video about “The Cloud”
• 9:25 – Bob Muglia is now onstage to discuss the Cloud…. but he’s a Server and Tools President.  I bet we’re going to talk about Private Cloud.
• Important keywords to describe a “cloud”: scalable, elastic, software, app platform, infrastructure.
• Discussing how “Bing” uses a technology called, “Autopilot” to manage their platform.  This was a “prototype” but not a really platform or product.
• Evolution of Computing
• 70’s – Mainframe
• 80’s – Client-Server
• 90’s – Web
• 2000 – SOA
• 2010+ – Cloud
• Cloud Application Model: service-Oriented, Always Available, Model-Driven, Scale-Out, Staged Production, Self-Service, Multi-Tenant, Federated, Elastic, Failure Resilient
• 9:35 – Don Box and Chris Anderson on stage!
• Discussing how Azure has evolved since PDC 2008
• Going to show off some C++ and CGI (not FASTcgi) coding running on Azure.  Guess we can’t take their word for it.
• Showing how T-SQL/TDS experience is available in Azure SQL (more than just the open REST enabled method of accessing data).  Including transaction enablement
• Shows OAuthWrap and OData feeds (Work done with Microsoft, Yahoo, Google). Using Javascript to authenticate and display data.  Very nice!
• 9:46 – Bob Muglia is back onstage
• Showing the relationship between on-premise and cloud services
• SQL Server – SQL Azure
• Windows Server – Windows Azure
• 9:47 – Kelly Blue Book dude showing their application using Windows Azure and Silverlight
• KBB has 14 million users per month.  Uses 2 datacenters for redundancy but often uses both for extra capacity
• Looking for a more flexible cost model….now going to use Windows Azure.  Application was already built on .NET and SQL.  To move to Azure, it required only a 1% code update.
• Uses 30 Azure Instances on average
• KBB is a “Data Provider”
• SQL Azure is used
• Vehicle data is published via an ETL process
• Community data is updated often and by users
• SQL Azure Data Sync is used to sync data between local and cloud instances
• Reduced time to provision new servers from 6 days to 6 minutes
• 9:53 – Video time!
• Dominoes Pizza uses Windows Azure
• Siemens using a Hybrid Cloud approach
• RiskMetrics Group enjoys computing on-demand
• 9:55 – Explaining how Microsoft is focused on providing connectivity between on-premise and cloud (Azure Only of course) services.
• Windows Azure, Code Name “Project Sydney” enables existing servers together with servers running on Windows Azure.  Available next year.
• Demo showing internal Microsoft application, “Giving Campaign”.  Deployed and running locally (corpnet) and uses Silverlight.
• Application running on Azure fails because the database is not available due to firewall
• From within SQL Admin, can create a secure and remote connection between the application on Azure and the internal database.
• 9:59 – Explaining the change in Windows azure Virtual Machine Role.  In the future, Microsoft will allow you to create and deploy your own predefined Virtual Machine.
• 10:01 – Announcing “Windows Server AppFabric” and is available today!  Cameron Skinner, Product Unit Manager of Visual Studio Team to show more
• Showing a simple “Car Rental” application
• Showing Multi-Monitor support in Visual Studio 2010
• Using ASP.NET MVC, Modeling.
• Showing single sign-on using Windows Federation
• Showing “InteliTrace” which shows you a trace of your application which has already ran and even takes you to the code for a specific function (see the SQL Query in the trace, click on it and take me to the code that ran it).
• Showing AppFabric – Enabled via Web.config and now handles all of the cache management.  Integrates within the IIS Managers and handles the cache, workflow, traces, etc. 
• Shows how to package and deploy your application via MSDeploy is “one-click”.  This is useful for internal and hosted applications, but it’s different than Windows Azure and the Web Role.
• 10:16 – Bob Muglia back on-stage
• Now talking about System Center.  System Center at PDC?  Interesting.
• System Center enables and manages the “Private Cloud” since it deploys and monitors your internal VM’s
• Douglas Purdy, Architect at Microsoft to show off early view into some System Center innovations
• Showing the “Application Model Project” in Visual Studio to model and define your application.  (I think this was project code name “Oslo”)
• Allows you to define your application components and dependencies.
• Shows Web Role, AppFabric Role, and Database Role defined and bound to the code that was previously created.
• Shows how easy it easy for a developer to “right-click” and deploy to Azure to test in a staging environment
• Shows how the developer can then take the tested application and create a “package” which is then deployed either to Azure or internally on Windows Server.  The IT guy can then use Powershell to deploy the application.
• Now showing how the application is being monitored using System Center Operations Manager
• 10:28 – Bob Muglia now talking about the Microsoft applications like SharePoint, Exchange, and Dynamics.
• Says how they will also build and leverage the cloud as well.
• Available Now: Windows Server 2008 R2, Windows Azure, SQL Azure
• Beta at PDC/Ship 2010: Windows Server AppFabric Beta 1, SQL Server 2008 R2, Visual Studio 2010, .NET Framework V4
• Beta 2010: System Center “Cloud”, Windows Server AppFabric Beta 2
• Datacenters in the US are ready today, others to be available in 2010
• 10:35 – Done!

Follow Me on Twitter 

Windows Live Writer used to create this post

Accelrys Pipeline Pilot Enterprise Server (PPES) is growing in usage within many scientific focused businesses.  I won’t go into all of the great things our product can do.  However, I am VERY focused on ensuring that we provide the right content, product updates, and tools to enable IT Professionals (ITPro) to plan, deploy, operation, and support PPES within their corporate environment.  Often times users and teams leverage PPES within their department to quickly analyze data and build interesting applications AND enable that data to be viewed and interacted with across the organization.  More importantly, it allows that data to get OUT of the lab and into the business (which could also be another lab or researcher somewhere).  With this kind of business impact, PPES typically moves beyond the department IT and moves to Corporate IT.  With that transition comes lots of questions and demand for more Enterprise IT capabilities.  This is where I come and my desire to build an ITPro community for PPES.

In the coming months, I’ll be working on building a community focused on ITpros so that I can better communicate and listen to feedback.  As you can tell, by reading this blog, I like social networking technologies like blogging, using twittter, and participating in a number of forums.  I’m hoping we can get a new ITPro section deployed on an Accelrys hosted forum, begin having a monthly ITPro update webcast, and whatever else the community (YOU) suggests.

If you’d like to participate, want to start giving me feedback (content, feature requests, etc.), or just want to say hello, please shoot me an email: cagramont (-A-T-) accelrys (-dot-) com  (sorry for the ugly email link….trying to prevent spam)

Follow Me on Twitter

This is my third week at Accelrys and now that I have a better understanding of the company and my duties, I thought it might be a good time to share that amongst the blogosphere.  First off, let me take care of some personal questions I’m sure I’ll get (really just saving myself from a bunch of emails).  While doing the consulting thing was fun, I really missed working with a team to deliver innovative solutions.  So Agramont Services as a business is slowly winding down.  Another key change is our family is relocating from Fredericksburg, Virginia to San Diego, California.  I’m JUST going to miss the East Coast Winter.  Nice!  But even better, we’re moving to where I grew up and have a ton of family.  For those of you whom I connect with via Facebook, you know of all of this already.  Lastly, Accelrys primarly focuses on Science which I have zero background in.  So what is an Enterprise/Hosting/Cloud/SaaS/Microsoft/.NET/Exchange/SharePoint/(you get the idea) guy doing at Accelrys? ….

OK, back to the business side…

From the Website: Accelrys helps organizations improve efficiencies through enterprise application of Scientific Business Intelligence software that can integrate, mine, model, analyze, manage and report complex scientific data.

I’m focused on Accelrys Pipeline Pilot which is Dataflow (different from workflow) which is both a platform for all of the other Accelrys products, but also platform for non-developers/knowledge workers/scientists to do powerful data integration, analysis, and reporting by simply drag and drop.  Pretty sweet!  Pipeline Pilot is growing in deployment size, scenarios, scale, and usage within an organization and is now requiring new features to support these scenarios.  This is where I come in.  I care about the Information Technology Professional (ITpro) who is responsible in planning, deploying, an operating Pipeline Pilot within Corporate IT or Business IT.  In the coming months, I’ll start posting information, questions, best practices, etc. both here on my blog, our corporate blog, and/or via the Accelrys online community.

If you have questions or thoughts on IT related requirements for Accelrys Pipeline Pilot, please shoot me an email (cagramont –at- accelrys.com) or connect with me via this blog or follow me on twitter.

In Part 1 of the “ISV Guidelines for Hosted Microsoft Dynamics CRM 4.0”, I covered the basics of leveraging Microsoft Dynamics CRM 4.0 (MSCRM4) as a hosted platform for Independent Software Vendors (ISV).  In this post, I’ll provide some guidance on the two most popular method of using MSCRM4 for an application.  This won’t be a deep technical dive, but general guidance.

• Part 1: Introduction
• Part 2: CRM as an Application or Platform
• Part 3: Shared or Virtualization Deployment & Licensing
• Part 4: Provisioning & Control Panels
• Part 5: Making the Leap

Part 2: CRM as an Application or Platform

All applications have some business value and problem to solve (ok, not games and stuff) and this is the main mission for ISV’s and the software they want to build and deliver to customers.  But before you can build it, you must first decide the platform in which you will deliver the solution.  This should NEVER start at the developer view (e.g. .NET or Java) or even decided on web or client application.  You must always start this at the core business problem (e.g. managing public records), define user roles, and define core scenarios.  Based on all of this, the list of features and technology platforms will come from there. 

OK, back to the CRM specific question….

Out of the box MSCRM4 provides a great (and typical) CRM experience. It has all of the standard structure and workflow that you would expect from a CRM application.  Under the covers, MSCRM4 stores all of the data in a SQL Server Database.  Hosted MSCRM4 deploys each new CRM instance as a separate database, but they all “roll up” to a single SQL Database which stores the configuration information.  This is what allows the “Multi-Tenant” capabilities of CRM.  Thus, a hoster can provides CRM services to multiple customers, leverage all of the same front (web servers) and back (SQL servers) end servers, and ensure that customers data are properly isolated.  I’ll discuss the provisioning of these servers and databases at a later time.  This is done using a feature of MSCRM4 called, “Internet Facing Deployment” (IFD).  Another key benefit to having IFD is enabling each client to have their own unique URL such as http://clientname.hosterurl.com/, but this can be remapped via a firewall/proxy (such as Microsoft ISA Server) to point to http://crm.clientname.com/, although it’s a bit tricky.  When a client connects to their services, you’ll get a form based web page to provide their credentials.  This is a generic UI and it’s the same for ALL Clients.  You “could” make this unique for your ISV application, but that won’t work in a shared environment as you’ll need to modify a few .aspx pages and .css files.  Making this change impacts everyone on the server.  In that case, you’ll need a virtualized deployment.  That’s Part 3 of this series.

Again, at it’s most standard deployment, it’s a great CRM platform.  But again, it’s generic.  This is where an ISV can provide some magic.  Now the question is, do you leverage CRM as an Application or a Platform.  By “Application”, I mean you can provide value to the CRM Experience by adding some additional Entities (more on that in a second) or workflow.  By “Platform”, you can leverage the structure of CRM and all of it’s functionality but abstract the entire User Interface for the application with your own.

First, let me cover some of the key areas of extensibility of MSCRM4.

• Entities – All “objects” such as a User, Organization, etc are called Entities.  There are many Entities that come standard with each deployment of a CRM Instance, but you can modify and remove those as well.  You can even create your own Entities that are specific to your application.  Each Entity has it’s own set of properties as well, which you can also add, modify, and remove.
• Relationships – Now that you have Entities customized, you’ll use “Relationships” to associate them with other Entities.  Let’s say you want to use a Standard Entity in MSCRM4 like User and associate it with an Entity of “Houses”.  This would be great for a Realtor application that wants to associate a collection of houses (Each house can have attributes defined like pool, rooms, etc.).  This can be customized per instance or even defined during the provisioning of a new instance (more on that in another post)

There’s WAY more to MSCRM4 then this, but this should give you a good idea on the flexibility of MSCRM4.  So you may ask, why not just do all of this in a database like SQL Server 2008?  That’s a great question, and sure you could do this.  But now you’ll have to develop that complete database structure, logic, web services (maybe), error handling, and more.  Developing all of that from scratch doesn’t buy you much and it removes you from focusing on adding the core business value which will make you a real profit.  Instead of throwing your development budget towards what MSCRM4, you could add additional features and scenarios that makes your application even more interesting to customers.

By leveraging MSCRM4 as an Application, an ISV can sell their products directly to customers that deploy CRM on-premise (e.g. they run their own server) or via a Hoster.  Since the hoster is selling Hosted MSCRM4 directly to a customer or via a reseller, the ISV’s component can be sold for an additional fee.  This opens the door to another issues in price modeling.  Since the customer isn’t purchasing their MSCRM4 implementation via the traditional perpetual method, but is instead doing it via a monthly subscription, an ISV must now take that pricing into account.  This is something you’ll need to work out with the hoster directly.

Now keep in mind that there will be some customers that will want to purchase your solution to enhance their Hosted MSCRM4 instance, but it may not reside on a hoster that you currently have a relationship with.  This could be a problem as the hoster may or may not give full rights of administration and customization to that customers.  There isn’t a standard here by the hosting community or by Microsoft.  On key customization component within MSCRM4 are “Plugins”.  A Plugin is used to provide a very rich customization point for ISV’s, but it’s not supported by many hosters in a shared platform because of the security risk it can pose (including Microsoft CRM Live).  Each Plugin can see and interact with ALL CRM traffic coming through the front end CRM servers.  Thus, even if a customer isn’t using a given Plugin, it will impact their overall performance and data.  This isn’t to say that a hoster won’t deploy it, but it will require a relationship with the hoster and allowing them to test and evaluate it.  That said, if an ISV chose to leverage the Plugin model for their MSCRM4 related product, this could be a deal breaker if a customer wants to deploy it in a shared CRM deployment and you don’t have a relationship with the hoster. I’ll discuss how leveraging a Virtualized MSCRM4 platform can help in all of this in a later post.

By leveraging MSCRM4 as a Platform, an ISV is going to leverage the structure and multi-tenant nature of MSCRM4 for the benefit of their application but remove the CRM-ish nature from their customer.  A good example of this is “Public Records Tracker” by Eskel Porter which leverages MSCRM4 to manage public record requests for State and Local Government.  The MSCRM4 platform is hosted for customers which reduces the deployment investments(hardware and configuration services), provides near instant-on access, and the User Interface has been customized to focus purely on the business problem.  Thus the customer is purchasing a subscription to an application to solve a business problem, and not purchasing a CRM solution with an add-on.

When using MSCRM4 as a Platform, there are two core scenarios to provide a User Experience to the end user.  First, you can fully customized the standard deployment of MSCRM4 via an XML template.  By doing this, you can still provide the basic MSCRM4 web user interface, but add/remove/modify the entire experience.  This reduces the development investment of recreating a User Interface (UI).  While this does provide a great deal of benefit from a time to market and development expense, you lose out in providing a customized and rich UI for the user.  Depending on your customer market, this could be a great tradeoff.

Here are some other areas of MSCRM4 that an ISV should know when customizing their solution:

1. Web Services (Site) – Each CRM Instance/Site, will have a unique URL to manage that given site.  This enables you to automate and integrate with the data structure (entities and relationships) and all settings for that specific instance.  We’ll talk more about this in a future section focused on provisioning.
1. http://<servername[:port]>/mscrmservices/2007/crmservice.asmx?WSDL&uniquename=organizationName
2. Web Services (Admin) – There is a specific management URL to connect to the web services exposed by MSCRM4 to manage that platform.  You can do almost anything, but you can’t provision/deprovision instances from it (bummer).
1. http://<servername>:<port>/MSCRMServices/2007/AD/CrmDiscoveryService.asmx?WSDL
3. ISV.Config – This is an XML file that allows an ISV predefine the entities, relationships, workflows, and UI extensions.  This makes the provisioning process a bit easier and can be automated by passing this information via a web service call.
4. Site Map – This is an XML file that allows you to fully control the navigation and structure of the site.  You can export this directly from within the MSCRM4 web application.  This is defined PER SITE.  This gives you lots of control, but does provides some headache if you plan on making adjustments to a collection of instances.
5. MSCRM4 SDK – There is an SDK that is available for MSCRM4.  It’s not required to customize MSCRM4, unless you plan on doing some heavy lifting customization by using Microsoft Visual Studio.  This is what’s needed to create “Plug-Ins” as we discussed already

Information on all of this can be found on Microsoft’s Developer Network (MSDN): http://msdn.microsoft.com/en-us/library/bb928212.aspx

Second, you can develop your application leveraging a very different development platform (E.g. ASP.NET, Java, Microsoft Azure) and “connect” to a given customers instance via a Web Service.  A powerful feature of MSCRM4 is the ability to fully interact with the data and logic via a Web Service.  By doing this, you can have full control of the application UI and thus removing all needs to the existing MSCRM4 UI.  The ISV can make all calls in the background.  You do this within the same hoster (e.g. CRM On Target is a great hoster for this) so your application AND CRM instances are within the same environment although perhaps not hosted on the same servers (More on that in another post).  The downside to this is that you must code EVERY piece of UI.  This can make for a better user experience for the customers by using Web 2.0 type of tools like JQuery, AJAX, JSON, and more while building it using ASP.NET/MVC.  Again, great for the end user, but it will drive up your development costs and time to market.

In summary, there are a number of ways an ISV can add value on top of MSCRM4.  And at it’s core, MSCRM4 will provide greater flexibility and business logic out of the box for an ISV.  The ISV must now decide if it will sell their product as enhancement to MSCRM4 or fully abstract MSCRM4 from users and use it as a platform for their application.

Note: If there are specific areas you’d like me to cover in future posts, please post a comment below or contact me here.

Follow Me on Twitter

Day 1 at the BizSpark Incubation Week was a great kickoff, but now the true fun begins!  I’ve decided to try something new and make today’s post and active log of our actions through out the day.  (We’ll see how good that works).  All times are US Eastern Standard Times

6:20 a.m – Wake UP and Commute – I live an hour south of Washington D.C. in Fredericksburg, VA and it will me about 90 minutes (I’m I’m lucky) to get to the Reston Microsoft Technology Center (MTC) where the event is held today.  Cross our figures!

8:00 a.m. – Arrived and Ready – The commute today wasn’t too bad and neither was the nice breakfast spread Microsoft provided.  Now I’m digging into the developer machine that Microsoft has provided in the lab for me to start some Microsoft Surface development.

An interesting note about Microsoft Surface Development is getting access to the SDK.  You may have seen the Microsoft Surface product before (or not), but it’s a really cool device/platform for creating an immersive, rich, and collaborative end user experience.  That said, in order to get a development machine (there’s a difference between the development and production device) it will cost you around $14k (I think the production one is $10k).  Only through purchasing at least a single development device will you get access to the Microsoft Surface SDK AND access to the developer community.  If you have a team of 5 people who will work on Surface development, you still only need the one machine.  The developers (those 5 people) will have access to an SDK that you install on top of your traditional development machine with Visual Studio 2008 (or Express Edition which is weird…if you can spend $14k for a Device, I’m sure you’ll have money for Visual Studio).  You’ll build out your Surface application using Visual Studio which includes a new project type/template of “Surface Application (WPF)” and “Surface Application (XNA)” installed.  For this week, we’ll focus on WPF.  The SDK also includes a cool simulator to run and debug your application on.  Even though Surface is a touch platform, you monitor doesn’t require (or will it even leverage) a touch screen.  It’s all done through mouse clicks, but the Simulator does a cool job of enabling you to mimic various user touches and even device simulations.

Now I start my first “Hello World” type project and see where that goes.  Today is really about learning how to develop Surface projects and then take a build and run it on the device.  Should be fun.

11:00 a.m. – Microsoft Surface Newbie – Now that I just started doing some Microsoft Surface development here are my notes and thoughts.

• Starting a new project (Yes…Hello World) was pretty ease.  Even adding videos and manipulating them was crazy easy!
• Make sure the Simulator is running when before you F5 to debug the application.  Visual Studio didn’t start it up for me.  (Nice to have feature I guess)
• Did some simple event handling (XAML for the UI and C# for the coding).  So far Surface development is very easy, straight forward, and with tons of benefits since you inherit many of the UI features like stretching objects (e.g. zooming into a photo), moving objects, and even inertia.

1:00 p.m. – Lunch and Guest Speaker – We had a guest speaker today by the name of “Loren Burnett” who gave a great presentation on “How to pitch your company to an investor”.  Loren was a very good speaker and gave lots of great tips on how to prepare, deliver, and follow up on your pitch to an investor.  I won’t give always his “secrets”, but was very helpful.  Especially on how to deal with Venture Capital funds.

If you have Questions or Comments today – Post it to Twitter for me (make sure to follow me first)!

3:00 p.m. – Moving off of Surface – I’ve reached a point where I’m comfortable with most of what we can do with Microsoft Surface, so now I’ll focus back on Windows 7.  We plan to leverage some off-shore development to build a new WPF application to manage some of the cool WPF Components that InterTouch Media is enhancing with the new Windows 7 Multi-touch.  The Intertouch Developer, James Cadd, has been doing some really cool updates with the code and we’re using an Imagiboard touch screen (who we never heard of) to do our testing.  So far so good.

10:00 p.m. – Wrapping Up – We made some pretty good progress and even had the official meal for developer gatherings… Beer!  We have one quick call with the off-shore developers which will give us our dream scenario…coding while we sleep.  Tomorrow will be all about kicking out more and more code.  It should be a long but fun day/night. I’ll have my Zune fully charged and ready to help me through the burn.  Night, Night all.

Follow Me on Twitter

Today I spent several hours in Reston, Virginia at the Microsoft office for the BizSpark Incubation Week focused on Windows 7.  BizSpark is a Microsoft program to help Startups build their software based application or service.  One of the chief reasons for doing this is enabling those startups to leverage the Microsoft platform and reduce their development costs by removing the up front fee for Microsoft Visual Studio via the MSDN Subscription.  That’s right, if you’re a startup company, Microsoft will give you all of their software for development for free! OK, not all for free as you have to pay a fee to get out of the program (after 3 years) which will cost you $300.  That’s crazy right?  Well, it reduces your risk and costs to get going.  If you’re successful in your business, you’ll pay Microsoft the standard cost for software when you exit the program (but now you’ll have revenue to pay for it).

OK, now to Incubation Week…

The purpose of the BizSpark Incubation Week is to bring together a select group of startups, pair them with an Advisor (I was one for this week—shameless plug), give them a lab, and provide some development resources.  For this Incubation Week the focus was on Windows 7 development.  The goal is to help startups learn and build some code that will leverage many of the new enhancements in Windows 7 (aka “Light Up”).  Training Content is based on material from the Windows 7 Training Kit for Developers.

Today was mostly an introduction and training (presentations, code samples, and demos) on Windows 7 for Developers.  It was a grind, but very useful.  Many in the crowd seemed very pleased and eager to get to the lab and get the code churning.

Here are some of the verticals of the BizSpark ISV’s attending the Incubation Week:

• Healthcare Security
• Real Estate
• Media Distribution
• Video Knowledge Management
• Mobile Search and Advertisement for Small Businesses
• Rich Multi-Touch Experience Platforms

Partners – These are the Microsoft partners that attended the event (and some were also advisors)

• Agramont Services – Based in the Washington D.C. area (Fredericksburg, Virginia to be specific)
• Vertigo – Based out of San Francisco area

Social Connections – These are some social networking information for the event:

• Use Twitter Hash: #iWk
• Facebook: http://bit.ly/9SMh2

Day 1 still isn’t over! And we don’t plan to end until around 9 or 10pm as we have some architecture and coding to do.  It's going to be a long week, but the people (Microsoft and ISV’s) are a great bunch and it looks like I might be able to do some Microsoft Surface development.  I’ll take pictures and post them (ok, no promises but I’ll try)

Intertouch Media is the company I’ll be advising this week.  They’ve done some wicked work with Windows Vista + WPF + Multi-Touch (here’s a video) to provide an interactive experience for touch screens.  They've focused on large touch displays and have some pretty big name customers.  Now we’re planning on porting to Windows 7 and Microsoft Surface.  The idea is to provide a multi-touch experience that spans and communicates between Windows 7 and Microsoft Surface platforms!

Should be a fun week

Follow Me on Twitter

A few years ago the United States Federal Government, specifically the U.S. Office of Management and Budget) created a PC standard for then entire government to follow.  The provided over 300 settings for Windows XP and Windows Vista in order to create a standard for all computers.  This is what is now knows as Federal Desktop Core Configuration (FDDC).  There is a ton of resources on the Internet, mostly on the .gov sites, that provides guidance on what these settings are and how to audit those settings using publicly available tools.

As with any IT Department, defining the policy is one major leap.  But to some degree, that’s the easy part.  Now you must deploy that configuration and ensure it stays enforced, not to mention audited and reported on.  With the U.S Government, having a mandate from the OMB is pretty powerful, thus making this problem space even more critical.

The FDCC is a perfect fit for Virtual Desktops from a deployment and management perspective.  Virtual Desktops is all about OS and Application standardization and consistency.  Thinking of having a pool of available OS instances, just waiting for a user to login from a remote device which could be a hardened thin-client or legacy PC.  All of those OS instances are based on a “Master Image” that has been fully configured with the FDCC policies.  When a user logs in, all of their applications are delivered via “Application Virtualization” (e.g. Microsoft App-V or Citrix XenApp) which is still abstracted from the underlying “Master Image”, thus keeping the desktop within FDCC standards.  All of the users data and application data is stored on a centralized store (e.g. SAN) which again keeps the “Master Image” clean of user data and provides additional benefits for the user and IT (e.g. daily backups of all user data).

So what about those users that go on the road?  Well this is where Virtual Desktop is still in play.  Using Microsoft MED-V or Citrix XenDesktop, a user can still take their FDCC approved image and applications on the road with them.  The bonus about Virtual Desktop deployments is that the process and image based deployments can be done directly on a physical machine as well.  You just take that master image, settings, and even application virtualization and deploy it directly on a laptop.  Using something like Microsoft System Center Configuration Manager and the Microsoft Deployment Toolkit (a solution accelerator) delivers this type of deployment scenario for both virtual and physical deployments.

Just like in any Virtual Desktop deployment, it’s not like Server Virtualization!  Managing the deployment and operations for a Virtual Desktop Infrastructure (VDI) is extremely different and requires lots of up front planning.  Not to say that Server Virtualization doesn’t, but when you consider the number of different users actually logging onto those Virtual Desktops, there are lots of end user scenarios you have to think through.  Even with the guidance of the OMB for FDCC (see, here comes the acronym soup), you may still define additional policies for given user roles.  Which could include access to applications via a variety of delivery models (e.g. web applications, application virtualization, etc.)

Follow Me on Twitter

Today on one of Citrix’s Blogs, they announced a new upcoming technology called “Hosted Virtual Machines” (HVM).  As if the Virtualization Soup of technology wasn’t big enough already, but this does solve an interesting problem.  Without much more information on the subject, here is my take.

Short Version: You want to host an application on a managed VM, but Terminal Services won’t work for a number of reasons.  With HVM, you use a client OS such Windows XP to run the application, but the presentation of it (just like Terminal Services) is then sent to the user.

Long Version: It seems that “Virtualization” is getting more and more attached to every new technology, but at the end of the day it’s about access to applications (that includes the OS and other applications).  Let’s put aside the delivery of an OS for now and focus just on the application.  There are a number of ways to provide a user with access to an application.

1. Traditional – This is where you get a CD or copy files from a file share and install the application locally
2. Terminal Services – Based on using a single OS instance, such as Windows Server 2008, and allowing multiple users to logon at one-time, but they each have their own “space”/desktop.  The display of that OS, or sometimes just a given application, is presented to the user.  Everything runs on the server, but show to the user on their computer
3. Application Virtualization – There are a few flavors of this.  This simplest view is about delivery.  The application is “preinstalled” and “captured” on a given OS (do a traditional install, but all files, registry settings, shortcuts, etc. are captured) and then deployed to any number of users.  So one “install” is then executed on any number of computers.  The application will run on the local computer, BUT it’s not installed there.  No files, registry, or shortcuts are anywhere to be found on your computer, but it still works locally.  That’s the virtual part.  Again, it’s all about deployment.

The big issue here is the ability to still provide “Terminal Services” like deployment of applications, but overcome some of the issues that “Terminal Services” (TS) has.  What kind of TS issues?  Well TS is still a Server OS.  It doesn’t have may of the client components (e.g. Windows 7) that some applications require.  TS is also Multi User based and there are some application that don’t work there either.

So why can’t Application Virtualization (e.g. Microsoft App-V or Citrix XenApp) work?  First off, there are certain applications that are developed either by a custom software development shop and built for a given customer and for a given OS/Application mix.  There are other applications that are certified by an Independent Software Vendor (ISV) that has specific requirements.  And then there are organizations like the Government, Health Care, and more that need to ensure that certain applications and data behave in a given way.  For all of these scenarios, an IT shop may want to provide an application to their users, but refrain from deploying them locally, it won’t work via TS, Application Virtualization won’t fulfill their requirements, and whatever else.

So the solution by Citrix XenApp (in the future) opens some very interesting doors.  I don’t think it will be part of the mass adoption, but it will break down certain barriers.

This leads me to think of other solutions such as Microsoft MED-V (Part of MDOP) and MokaFive that provide this kind of host based virtualization, although with HVM Citrix also allows this to be hosted on a server.  I guess I’ll have to wait for a Citrix demo and trial for me to learn more.

BTW, I wonder how this will impact hosters looking to get into the application delivery model.  Since this does require another client OS, Citrix rightfully notes that you’ll need the Microsoft VECD license.  Too bad VECD is not on the Microsoft SPLA list.  Bummer

Follow Me on Twitter

As it stands today, there is a large and growing Virtual Desktop Infrastructure (VDI) market with vendors spread throughout the software and hardware spectrum.  Unfortunately, many enterprises have either not done or heard of VDI or their in perpetual piloting.  My theory isn’t that the benefits of VDI isn’t valid, and it’s certainly not for a lack of value realization once a solution is deployed.  My take is that since many organizations skipped Windows Vista and stayed with Windows XP, they haven’t had much motivation to make the move.  Sure, there are tons of benefits (power, management, centralized storage, etc.) to VDI, but it does require rethinking how IT plans and deploys desktops.  To be frank, it can be a huge headache regardless of the value.

Server Virtualization, however, doesn’t have that same pain in deployment.  Yes, you need good management, deployment strategies, etc.  But to just try it out, you install the Hypervisor (e.g. Microsoft Hyper-V) and then deploy a server OS as a virtual machine on the same box.  There, a pilot!!  You might think it’s the same for VDI as you just replace that server OS with a client OS (E.g. Windows Vista), but there is the rub.  With server virtualization, you typically deploy a give server OS version (e.g. Windows Server 2008 R2), deploy a server application (E.g. Exchange Server 2007), and then you’re off.  With VDI, you may want to deploy a Pool of clients OS’s (e.g. Windows Vista) based on a given desktop image (they are the same, but are unique in computer name), use application virtualizations to dynamically provide applications to users, and centrally store user data which is abstracted from the client OS (E.g. not stored in the C: drive, but stored on a Storage Server or SAN).  Building out this type of environment takes good planning, testing, and deployment to ensure end-users get an environment that “feels” like a typical desktop, but is policy based, dynamic, and managed as a collection.

Ok, so why will Windows 7 make this all worth the pain?  I won’t go into the end-user features of why Windows 7 rocks (that’s for a later post), but Windows XP is showing its age and so is the hardware that end-users are running it on.  So now an IT department must think about what to do with all of that old hardware and how to get Windows 7 to all of their users.  There is still the traditional option of purchasing new desktop/laptop hardware and migrate user data one by one.  This is typically a lengthy and painful process.  Plus there’s no turning back!  With VDI, an IT department can target which users get access to the new environment based on Active Directory security groups, migrate end-user content weeks in advance, allow end-users to use both legacy (what they have today) and Windows 7 (using VDI) at the same time.  Plus, you can even provide access to Windows XP, Windows Vista, and Windows 7 desktops (With access to all of their own data) VDI instances from their same desktop for application compatibility support.

Oh, and they can do all of this using their existing desktop as a “thin-client”.  Like I said before, there are many moving parts to make this work including licensing (which Microsoft has made huge improvements on), so be sure to run a pilot first to get a feel for how it all works BEFORE you do heaving planning.

So in summary, due to end-user interest in Windows 7 and the aging end-user machines, IT departments will evaluate and deploy VDI solutions within the coming months with greater zeal than in the past.

Follow Me on Twitter

This is SWEET news!  Windows 7 and Windows Server 2008 R2 is now Released To Manufacturing (RTM).  This means that Microsoft is done with the code, sending it to various Original Equipment Manufactures (OEM) such as Dell, HP, and IBM. Everything else for Windows 7 will be done via patches and Service Packs (as usual).

I first saw this news via Twitter (@MSWindows) where the Windows team then blogged about it:

http://windowsteamblog.com/blogs/windows7/archive/2009/07/22/windows-7-has-been-released-to-manufacturing.aspx#commentform 

So what’s next?  Next you’ll see it on MSDN and Technet within the coming days and then customers with an Enterprise Agreement (EA) or Service Provider License Agreement (SPLA) will see it next.  For everyone else,  it will be available on October 22nd.  Will we see people around the block ready to purchase it at Midnight?  Perhaps a few, but I think the magic will REALLY happen during the Holiday shopping season with Netbooks.  You’ll get a sweet new Netbook between $150-$300 (US) and be running the latest from Redmond…Windows 7!  When more people have Windows 7 in their hands thanks to Netbook, I predict massive adoption during the second half of 2010.  Just in time for Office 2010!

For those interested into getting their hands dirty and ready for Windows 7 (here comes the plug), checkout my Windows 7 Services (including Virtual Desktops)

http://agramontservices.com/Desktops.aspx

Follow Me on Twitter

I had a great call the other day with the CTO of Pano Logic, Aly Orady.  As some of you know, I’ve been doing a fair amount of researching and lab work with various Virtual Desktop Infrastructure (VDI) technologies.  In looking around for VDI solutions, I came across Pano Logic which is based in Menlo Park, CA.

Pano Logic provides a unique spin on the VDI space as they’re solution encompasses both the software and hardware component.  Sure, many hardware vendors, such as ChipPC and Wyse, offer a software component as part of their hardware solution as well. But there difference is that in order to use the ultra-sexy device from Pano Logic, you MUST use their software component as well.  The reason their software is required is because of how the device actually communicates with the associated Virtual Machine.

The majority of “thin-clients” share the following characteristics:

1. Small Form-Factor – Since all of the “real” computing power, data, etc. is done on the server, there is little requirement for a typical PC footprint.
2. Embedded OS – These devices will normally leverage an operating system to handle access to the peripherals, network connection, and the protocol interaction with the server (e.g. RDP for Microsoft, ICA for Citrix).  In a Server-Based VDI scenario, the server handles all of the processing. The client is then sent screen updates which the device then presents on the attached monitor.  Embedded OS’s include Windows CE, Windows XP Embedded, Linux.
3. Device Management – When you have a number of thin-clients deployed, many times you need to keep track of them, provide software or firmware updates, and provide other levels of services for those devices.  If you have a management platform (e.g. Microsoft System Center Operations Manager & Configuration Manager), many times you can manage and monitor those devices using that platform.

The difference with the Pano Logic solution is that device does NOT include any software based OS on the device.  How can that be?  Their device functions more like a true dumb-terminal (not to sound mean or anything).  It passes all I/O commands from the device to the Virtual Machine where your desktop is being hosted.  The idea is that an Embedded OS on the device requires updates and is vulnerable to security risks. 

When the device is powered on, it looks for a “Pano Manager” which is the broker between the device an the target Virtual Desktop.  Pano Manager is included as part of the solution, so you don’t need any special license for that.  It also isn’t a single point of failure, because it’s no longer needed once the device is connected to the Virtual Machine.  The Pano Manager is deployed as a “Virtual Device” as it’s ready for production, built on top of the OBF format, and wizard driven to make deployment time 30-60 minutes.  Within the Virtual Machine (VM), there are special device drivers that enables the communications between the VM and the Pano Logic device.

The Pano Logic solution today leverages a VMware infrastructure to enable the entire VDI solution.  You can leverage ESXi and Pano Logic, but you won’t get the fully automated deployment you would get with VI3 + vCenter from VMware.

Pano Logic is a cool solution “within the firewall”.  Since the device isn’t too smart, it can’t be used from a remote location (e.g. a telecommuting worker) since the device needs to find a Pano Manager on the local network.  You could work around this with a point to point VPN solution, but this would have to be done on a separate device than the Pano client (e.g. a firewall/router with VPN capabilities).

The devices are sold via their Partner Channel and their current sweet spot of customers are Mid-Market Enterprises and Government ranging from 5-10k clients.  You only need to purchase the client device/license which will give you rights to any number of Pano Manager’s needed to meet the architecture and scale of your deployment.

Other points of interest:

• Multi-Monitor Support: No/Yes.  A single device can not support multiple monitors.  However, you can “bridge” multiple devices to enable this behavior
• DVI Support: No
• Microsoft Integration (they are looking into this, but no official roadmap just yet)
• System Center: No
• Hyper-V: No
• Device
• Designed and built by Pano Logic
• Ports for: VGA, Ethernet, sound, microphone, 3 USB ports
• 2.5 Watts per hour
• Pano Manager
• Integrates with Active Director
• Integrates with VMware vCenter
• Integrates with Third-Party brokers

With the solution from Pano Logic, an enterprise can quickly get into the VDI game and they don’t have to worry about many headaches in deploying, configuring, and tracking their devices.  That said, there are still other challenges that are really required for a robust VDI solution (e.g. application virtualization, user data centralization, remote users, mobile users, etc.).

I haven’t had the chance to play with their software or device just yet, but I’ll be sure to blog it as soon as I do.

Follow Me on Twitter