Community
Post of the Day
November 5, 1999

Board Name:
Akamai Technologies

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light.

Subject:  Re: The Competition
Author:  TMFDwight

Hey Mark,

that would be me
Wow - very cool. Thanks for posting.

where do you weigh in on the technology argument?

You and many others on this board don't want to hear this. I think Akamai totally rocks. I have been working on a couple of lengthy posts for this board but have not had the time to finish them. The demands of work and being a new father have put me pretty far behind. Following are several points that I believe to be true. I'll post follow up messages (probably this weekend, if I can get any free time) going into more detail about all of them. So please put down the flame throwers until you read the follow-ups :). My points:

- Akamai does have a sustainable competitive advantage in its algorithms.

- It would be very difficult for a competitor to do what Akamai does.

- We have found Akamaized content to be significantly faster (at times 8 to 10 times faster) than non-Akamaized content.

Why Doesnt the Fool just rent a server from Exodus and putyour graphics on it and save some money.

This would do nothing for us or any other content provider. My customers are not just on Exodus or AboveNet or Frontier Global Center or any other ISP. My customers are on hundreds of different ISPs around the world. Making our content really fast in one or two ISPs networks is not gonna do the job.

"The folks at Yahoo are totally hard core and unbelievably smart. They do pretty much everything themselves."

I have been doing the Fool tech gig for over five years. In that time I have seen a lot of changes. It used to be that going through the MAE's was painful but private peers were fine. That does not seem to be the case any longer. Even private peers are sucking wind. When we see problems, we contact the ISPs. Generally we get two companies pointing fingers ("can't be me - must be the other ISP.") but not much help. The result is that any time traffic has to traverse two networks, it's generally gonna suck. Akamai helps avoid this problem by placing my content in many networks. InterNAP is another company that we are looking at that can help us avoid this problem.

This is one of the main reasons I like Akamai. They have thousands of servers in many networks. So there is a very good chance that my customers will be close to an Akamai server and will not have to go through a public or private peer to reach the server.

True - they do not own their own bandwidth. I don't see that as a problem. That is one less huge problem to manage. Bandwidth is becoming a commodity and Akamai is providing a value-added service for which I and some of the largest content providers (including Yahoo - the company that owns Broadcast.com) are willing to pay.

save some money

Mark, with all due respect (which is a hell of a lot), you're kidding, right? Save money by putting a few servers in several networks? That means I have to pay for the servers, the rack space, the bandwidth, and the people to monitor and manage the servers. That's expensive. Or I can pay Akamai for the bandwidth. Granted, I do pay more for the Akamaized bandwidth than I do for bandwidth from an ISP. However, it is totally worth it. The benefits are three fold:

1. Lower infrastructure costs. Let's look at a random day last week - Thursday October 28th. On that day Akamai served over 12 million hits for Fool.com. Those are hits that I didn't have to serve with my servers. That means the load on my servers is lower allowing me to stretch them further. This lowers my hardware and management costs.

2. Lower bandwidth costs. Remember those hits I am not serving? They are the most expensive ones - my graphics files. I don't have to worry about (or pay) my ISP for that bandwidth though I do have to pay Akamai. The result is that the load on my routers, switches, firewalls, and load balancers is much lower. Again - lower hardware and management costs. One interesting note - when we rolled out Akamai across the Fool site, we saw about an 80% drop in bandwidth utilization. That's pretty cool.

3. Better customer experience. As I mentioned above, we have seen improved speed with Akamai. YMMV. I have seen a few customers complain about Akamai. Several of the complaints were from AOL users. I won't go there as my "issues" with AOL's web experience are many. I will say that if 100 customers out of a couple of million complain, well � that's a pretty small percentage. Of course, I would like for everyone to be pleased with the customer experience at the Fool but I realize that we cannot please all the people all the time.

"Add it up and Akamai is pretty compelling from a content provider's point of view."

Add it up and Akamai is pretty compelling from a content provider's point of view. And clearly I am not the only one who thinks so. There a few other pretty big players who agree with me: Go.com (InfoSeek), Apple.com, About.com, The New York Times, The Washington Post, CNN.com, and Yahoo, just to name a few. Would all these companies be using Akamai if there were nothing there? I kinda doubt it.

The really interesting one, at least in my little mind, is Yahoo. The folks at Yahoo are totally hard core and unbelievably smart. They do pretty much everything themselves. If they can't do it themselves, they buy it (like Broadcast.com). The joke around the Motley Fool TechDome is that Yahoo smelts its own metal to make its own servers. They are that self-sufficient, well ... almost. Yet they are using Akamai. They didn't build it themselves. They are outsourcing. That speaks volumes, at least to me.

OK - I've got to run so I should quit rambling and sign off here. As I mentioned, I'll post a few follow up messages that go into more detail on the points above. Before I go, I should post a -

DISCLAIMER: I am long AKAM. Take all of the above with a boulder of salt if you choose. I will say that I have been bullish on Akamai since before they went public. And I bought the stock because I believe in the company. I do not believe in the company simply because I bought the stock.

Fool On,

Dwight