Tag Archive | streaming

Any app, from the cloud, on any browser.

Alright I admit it, I’m jealous.  I joined a start-up! I’ve seen Silicon Valley! We were going to change the world, I was going to be rich beyond the dreams of avarice, leave the rat race behind and open a beach bar somewhere. But you’ll have guessed by the fact that I’m writing this blog that that never happened. With hindsight, I would have joined Frame, the (fairly) new face of cloud-hosted application delivery. Their premise is simple; run any Windows application in the cloud and access it via a browser, no plugins required.

Originally called MainFrame2, the company began life enabling ISVs to offer applications as a service. It got off to a good start but its fortunes improved massively when the focus changed to end users and the business was relaunched as Frame. With recent investments from Microsoft Ventures, Bain Capital Ventures and In-Q-Tel growth continues at pace. On top of that they recently signed a major partnership with VMware to become part of their Workspace One offering with App Express.


Frame is essentially an Application-as-a-Service company, built for the cloud in the cloud. You install the applications into a sandbox environment and then, when you are ready, publish them to the Frame Desktop (as above) for users to consume. Your applications are installed onto Windows 2012 servers (the roadmap is for Windows 2016 and 10 soon) with the ability to make use of the GPUs offered by AWS and Azure to handle even the most intensive graphical applications. Those screen images are then delivered by Frame’s encrypted and highly compressed display-protocol to the end user allowing any application to run on almost any computer. Removing the complexity usually associated with virtual desktop computing to a few clicks.

So what are the uses for technology like this? Here are a few examples:

  • Think about those expensive CAD and desktop publishing packages. With Frame you can centralise them in the public cloud of your choice, share the licensing costs, utilise cloud storage to make collaboration easy and reduce the need for expensive workstation hardware*
  • Consider the education sector and the ability to use inexpensive Chromebooks to access any type of application and then not having to pay for those resources during the holidays
  • Mobilise legacy business applications by migrating them to the cloud and using Frame to provide browser-based access without having to install anything on the client
* and not just hardware as Microsoft have brought in a new Windows 10 Pro for Workstation licence that affects any machine that has an Intel Xeon or AMD Opteron processor.

However, Frame is not for everyone or every use case. It’s not going to be a way to deal with legacy applications to aid that Windows 10 migration. If it won’t install on Windows Server 2012 it isn’t going to work. You also need to understand your responsibilities as a customer. Although you don’t need to licence the OS you still need to patch it, supply your own anti-virus client, update those applications and then secure the network access to it. And don’t think you can escape the fun that is Evergreen!

Cost-wise there’s a $ per month, per-user charge based on standard, pro or enterprise levels of functionality. Then an hourly rate based on usage and the resources that your VMs consume. Automation is key to controlling those costs ensuring that machines are not costing you money when they aren’t being used. There are features within the administrative console and the REST API to schedule the number of machines available and for those machines to be powered off when they aren’t required. Calculating the overall cost, like a lot of cloud initiatives, is not an easy one though and may not be necessarily cheaper than your current on premises solution. But there are features and functionality that no on premises solution will ever give you.

The big differentiator for Frame is its simplicity and ease of use. When you need to bring additional services you just plug them in. You need identity services? Frame supports them. You want to use your user profile management tool? No problem. Want to connect to Dropbox, Box or Google Drive? A couple of clicks and it’s setup, appearing as a mapped drive within the Frame explorer. Want to share your session with someone else to work on a document or drawing simply email them a link to the session? Need additional local storage or a database? Just click the utility server option and select your services.

Just as data and business applications are moving to the cloud, it makes sense for client applications to follow them. Another nice thing about Frame is that where companies utilise multiple clouds you have the ability to place your applications in the best location to serve them avoiding any lock-in. Also, as client estates become more diverse and the demand from users to work from anywhere increases so the ability to deliver applications simply through a browser becomes increasingly enticing.

Frame is very cool technology. If you’re currently considering XenApp running in Azure or XenApp Essentials, or considering at how to mobilise those legacy applications, then you need to take a look. There are limitations as to where it fits as a solution but where it is right there are clear benefits. Frame enables powerful applications to be accessed from almost any device. It enables applications to be delivered to an entire business anywhere in the world minutes after installing it once, regardless of the endpoint they are using.

So my dalliance with the world of start-ups was not a great success. For the guys at Frame I can see a much brighter future. The question though is how long will it last before someone swallows them up?

The beginning of the wrong end – dare we consider the impact of a “multi-tier / multi speed” Internet?

One of the most fundamental pieces of news for ALL Internet users broke last week across in the USA but seems to have slipped under the radar over here.

Put simply it’s the start of a change of stance by a number of major US carriers of “data” to levy additional charges to content providers that generate large volumes of Internet traffic. To explain this further, at present if an end user chooses to use the service of “content provider A” they access it via whatever internet connection or point of presence choose to use (whether a paid or free service), fixed or mobile. But the current result could be a popular service (for example social network site content or video streaming) delivering a mass of internet traffic generated by “content provider A” across the major carriers networks with no additional charge paid to the carrier (who must still maintain quality of service, manage bottlenecks, etc.).

However the news broadcast last week highlighted a change of stance by a major US carrier who is now requesting an additional charge from a major TV/film streaming company to carry its traffic across the carriers’ network.  And what happens if the content provider refuses to or does not have a cost model that supports the payment of such a charge – does that mean traffic generated by an end user of the service is discarded, rejected?

This overt change seems to be by many as a first small step to a multi-tiered internet, not the “free ish” flowing internet we have today. And the worry, what originates in the US seems to have a tendency to quickly permeate to the UK/Europe (and how could this not).

I revisit the title of this blog, “Could this be the beginning of the wrong end”, which sees types of content only running at optimum levels when transported via “Carrier As” network but not “Carrier B”? And what happens when the networks join at various parts of the Internet, will traffic formally slow down at certain points because “Content provider C” hasn’t paid the carrier “traffic transport” premium?  As an end user of an internet service does anyone really need to understand where, who and how to connect to gain not only the best experience but potentially the service at all?

This could be deemed an unsolvable problem as many think for carrier to seek to maximise the monetary income from transporting data is not unreasonable especially when they are fundamental to service delivery – but equally it’s tough for content providers (and that is virtually everyone on the internet) to factor in yet another variable in their income cost model (if they have a cost model at all).  The Internet is only of value if “content” is available to the widest audience and can deliver the optimum end user experience with/from that content – a variable end user experience without the end user understanding why does not bode well.

This one is one to watch with very interesting times ahead.  However this is resolved, and there are unlikely to be any true winners, ramifications to every popular content provider on the internet are great (and likely to cascade down to the end user).

Watch this space.

Until next time.

Colin W

Twitter @colinwccuk