Amazon’s CloudFront Now Offers Flash Streaming, This Will Disrupt The Market

This morning, Amazon announced that their CloudFront content delivery service now supports Flash streaming as long as the videos are stored on Amazon's S3 service. Amazon has deployed version 3.5.2 of Adobe's Flash Media Servers in 14 edge locations globally and now supports RTMP, RTMPT (HTTP tunneled), RTMPE (encrypted), and RTMPTE (tunneled and encrypted) flavors of RTMP. In addition, Amazon says they will support live Flash streaming sometime in 2010.

While Amazon's CloudFront service for on-demand video never really disrupted any of the larger CDNs in the market, this latest announcement drastically changes things. While Amazon is not going to take any large customers away from the CDNs, think MLB or Hulu, Amazon does have the potential to take many of the mid-sized customers who spend between $3-5k a month on video. That's Amazon's sweet spot and there are a lot of those customers in the market with the major CDNs. In addition, while Amazon still needs to flush out the other ecosystem pieces for video, their offering already ties in with transcoding solutions from and Bits on the Run. Additional third party developers are already working on other apps that will help enhance Amazon's offering including reporting.

While some might say that the downside to CloudFront is that it does not have an SLA, Amazon does have two levels of premium support that costs between $100-$400 a month for customers who want 27×7 phone support. Right now, I don't see the lack of an SLA for CloudFront as a negative for the customers they are targeting. The trend I am seeing in the market is that content owners want more control over their content, want to be able to do self provisioning and don't want to have to call their CDN every time they need something done. Amazon requires no commitment, no contract, no overages and has up-front pricing with no additional cost for Flash streaming. You can setup a CloudFront account right now and be streaming in about five minutes. What's not to like? If some still want to argue that delivering video on the Internet is not a commodity, Amazon's announcement should make them think again.

By far, the biggest impact this will have is on the smaller CDNs and streaming media hosting providers. While the smaller providers will still offer more hand-holding, better support and help with things like building custom players, Amazon will add those options as well over time, working with third party developers to build apps on top of CloudFront. As for as CloudFront's effects on pricing in the market, I don't see it having any impact at all. I know some will suggest that Amazon's Flash streaming service will drive prices down, but Amazon's pricing has always been published on their website and in a call with them this morning, Amazon said they don't plan to change their pricing for streaming delivery.

The only real question in my mind about Amazon's new Flash streaming
service is how it performs. Amazon said they don't do any benchmarking testing against any competitors but felt confident that after running CloudFront for over a year, customers would be very happy with their Flash streaming performance. In many cases, Amazon won't need to be better than the leading
CDNs. As long as they are on par, or close to their competitors
performance, they will do very well with the solution. I'll do some real testing myself on the performance of the service in the New Year.

Amazon's future road map for CloudFront is all based on requests they get from customers and when I asked about the possibly of Amazon supporting Sliverlight soon, Amazon said customers simply weren't asking for the format yet. I also asked Amazon why they picked Flash Media Server over Wowza, who has a good relationship with Amazon, and again, Amazon said customers were specifically asking for support of FMS 3.5.

In the CDN space, vendors know who the top 100 customers are of their competitors and are constantly knocking on their door. While Amazon does not have the same strategy, the advantage they have is that every day, tons of customers sign up for CloudFront services through their website. Amazon has a very large built in base of customers they can up-sell and the company did say that account managers do reach out to the largest content owners on their network.

While I don't see Amazon changing the CDN landscape over night, they are already starting to have an impact on the market and as they continue to add more functionality to CloudFront, their impact will only continue to grow.

  • It will be a battle of customer service and service level performance to be sure… Self provisioning alone is not a winning formula when it comes to media delivery- even at the $3K-$5K range.
    $5K per month buys a lot of bandwidth these days, and I’m not so sure that self-service is a guaranteed winner in this segment of the content delivery world. $5K on other CDNs, for example, buys significant international connectivity and presence…
    But time will indeed tell… I’m glad to see more competition.

  • Keep in mind that Amazon FMS Cloud is based on the FMSS version (Flash Media Streaming Server) which is limited to streaming video and not the full FMIS version (Flash Media Interactive Server) which supports the C++ plug-in framework, and the server-side action script framework required for developing custom streaming solutions and multiway communication services.

  • Charles

    This is more disruptive to the INTERNATIONAL markets more so than the United States. Akamai, CD Networks and LLNW now have a ~.20/GB price point to match in EMEA & APAC.

  • Tanner Boyle

    So what percentage of the big CDN’s revenue is in this sweet spot? Or more importantly what is the margin on the lower end of the CDN customer spectrum?
    Is Amazon disrupting Limelight’s couch change or is this 25% of their gross profit?

  • Good question Tanner, all depends on the CDN. Someone like EdgeCast, this is their sweet spot. Someone like Akamai and Limelight, hard to know exactly but I would guess 35% of their business is in this contract value range, but that does not mean the services being offered are the same.
    Right now, Amazon is not disrupting any of the bigger guys, but they will over time.

  • Personally I love the self-serve solution that Amazon offers and we have experienced great success in testing it. I can see a lot of opportunities for development teams such as us at Eyepartner who builds widgets and HD broadcasting API’s as add-on’s to the Cloud.
    Too often when we have approached CDN’s in the past, with our HD Broadcasting software only to be told that they did not want to be involved in the software distribution business. It’s amazing that the CDN’s fail to see the significance and ROI of offering a turnkey LIVE HD broadcasting play.
    With this self serve platform with and no minimums Amazon will be able to offer their services so that businesses can try their hand in broadcasting out without a large upfront infrastructure cost. This is the reason that I think Amazon will take great deal of business from the CDN’s.

  • It seems like Amazon CloudFront has taken its place on the video streaming market. I’m not sure if they will be able to really focus on this type of service in the future considering the whole bunch of completely different services they’re running!