Archive for the ‘ISP’s’ Category:
There is no European market
U.S. based CDN’s often tend to think that they can enter the European market by having an office in London, Paris or Amsterdam.
Although there is a European Union, without borders, with free economics and a single currency, there is no single market like there is in the U.S.A.
There are over 25 countries and even more languages in Europe. You will encounter cultural and language barriers. Opening an office in London also means a geographical barrier.
Why are telecom operators deploying CDNs?
Telecom operators have a four-way strategy to own an on-net CDN. On-net CDNs offer deeper network penetration with better QoS and larger capacity. The CDNs that we deploy support all popular delivery technologies for the web, mobile and IPTV. This means that the telecom operators can:
- optimize on-net traffic flow, reduce backbone and peering load
- host and deliver on-net web, mobile and IPTV services
- sell CDN resources to content owners
- Take back the distribution role in the value chain
External CDNs could only have addressed number 3 in this list. Read more »
Peer2Peer s*cks, here’s why
Peer2Peer (P2P) is sometimes proposed as an alternative for distributed content delivery. Some CDN’s are built upon P2P technology. P2P vendors, researchers and CDN’s claim that P2P lowers costs. It does. For the content owner. But it raises costs for the network owners. Significantly.
Basically the concept of P2P is that every client can redistribute content to other clients. As a content provider you don’t need to buy servers or traffic. Your audience distributes your content ‘for free’. Ingest the content into the network and let it flow. It is also hard to trace where content is hosted and ingested from. That is why P2P is so popular for (illegal) file sharing services.
Digital Logistics
2002. Since I already had a background in distributed delivery, I dived deeper into the internet infrastructure, to come up with a smart distributed solution for mass-scale content delivery.
It involved a simple number of technologies I already had developed in operational or rudimentary form. Parallelized delivery servers as core and edge servers. A smart asset and live relay replication mechanism. A smart geo load balancing application. Some log processing scripts.
Technical stuff. But why and how does this work? Let’s compare this with Logistics. Read more »
Video Working Group
2002. I had done a lot of research on broadband capacity, edge distribution and CDN management. I foresaw that the Internet would break under the increasing load of online video. I had experienced (eh caused) it myself one time.
The ISP’s would struggle to keep their business case positive. Content owners would be forced to outsource delivery to CDN’s. Who would only be able to distribute to ISP’s.
The Connected Home
2004. Ah, Fiber to the Home. The holy grail. According to many. Experts warned the government that if we did not invest in the Last Mile today, the Netherlands would become a third world country and internet services would never take off. So they asked the government to invest millions of tax payers money in new Last Miles.
I was a bit sceptical about their claims. Because I coincidentally had done some research about the bottlenecks of the Dutch Internet. And my conclusion was that the Last Mile could perfectly handle future growth, but that the true bottlenecks were at the ISP core backbones, the Internet Exchange and their interconnects (public and private peers).
The experts claimed that broadband services would require more than 20Mbps capacity per household. Maybe even 100Mbps in the future. But there was no research done to prove this claim.
Open Play
In 2001 I came in contact with the Dutch Innovation Platform, from the Ministry of Economic Affairs. Because I foresaw a major shift in the value chain.
Thanks to the Internet, consumers were not locked anymore in the old walled garden, linear value chain:
Consumer -> buys television package from cable operator -> buys tv channels distribution rights -> broadcasters add advertising revenues -> pay the content producer.
The open Internet allowed multiple new value chain models:
Consumer -> buys flat-fee broadband. End of chain.
Consumer -> surfs the web for content aggregators who ask money for their service or add advertising revenues -> to pay the content producer.
I called this Open Play. Read more »
Sorry
2000. We produced a large webcast for the EuroSonic / Noorderslag pop festival. We had built another ad-hoc CDN with streaming servers in multiple Dutch ISP networks.
Each server was connected to 1Gbps. But one of the ISP’s didn’t tell us that that was the capacity of their entire backbone. So we broke the Internet for many, many users that night.
They couldn’t get their email. Or surf the web. But we had many enthusiastic viewers. Sorry! :-)
So much for a CDN, right? We needed to talk to the ISP’s because if their networks couldn’t handle the streams, how could we grow our webcast production business? So that’s what I did.
