Slashdot Traffic, Site Statistics and dasBlog Performance
I was talking with Omar today and he wanted to see some traffic details posted from the day I was Slashdotted. Some folks didn't think that dasBlog was up for the test. Frankly I was surprised that the traffic was so low! Seems like everyone's site tips over when they get Slashdotted. I'm just running the standard shared hosting plan at OrcsWeb. Even during the height of it I was still running around in the admin interface without a burp. I was getting quick page views and hadn't even turned HttpCompression on yet. Note, this is just page views, not every Http GET (don't want to count gifs, etc) or Aggregator Traffic.
| Date | Views |
| Sun Dec 21st, 2003 | 1,751 |
| Mon Dec 22nd, 2003 | 12,928 |
| Tues Dec 23rd, 2003 | 4,227 |
It doesn't seem like that much traffic to me. I transfered 1.54GB that day.
Then again, I only get about ~2,000 page views a day; most of my traffic is Aggregator traffic. Monday's have typically been my biggest days.
Why did dasBlog make it and others fail? Maybe this says something about the average web site on the net? Or the average hardware? Or .NET?
Or maybe it's just that with the proliferation of Blogs out there, more people get there news from the set of all sites that are NOT Slashdot?
What do you think?
About Scott
Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.
About Newsletter
Some of the slashdotted sites I've seen that crash are simple HTML files, so this is very interesting.
BTW, great job on writing interesting, unique and controversial content that got you slashdotted in the first place!
I'll look for traffic bursts for that day. Perhaps it's the bursts that knock sites over. Most of the traffic came steadily between 9am and noon PST.
According to the event logs, I had only 2 errors that day, which were the "collection failed to enumerate errors" that can happen if you change content while it's being eneumerated. But noone was turned away, and ASP.NET didn't recycle.
It's usually the spike that gets people. But given that this topic (new ways of filtering email) had been covered a few days earlier, most people probably didn't read your article. Just said the same thing they had earlier or responded with the knee jerk response "Wow, threaded views in email. Just like technology {0}",InsertOldTechnologyHere
http://www.mcs.vuw.ac.nz/~kirk/#20021025
I don't have the figures, but from the pictures it looks like more than a gig of traffic for each of the first two hours. Ours was an html page linked to a 2mb PDF download, and what temporarily maxed out our server was too many open connections - because it took too long for each person to download the PDF, the connections were open too long.
Comments are closed.
Disclaimer: The opinions expressed herein are my own personal opinions and do not represent my employer's view in any way.
Let's see. 13K page view over 24 hours means average of 9~10 page views per minute. I don't know how many images your page had but assuming 20, you got 3 resource requests per second. I would say that was pretty good particularly since hit rate would have been pretty high at peak time. Did you have front page cache on?
Also there is the question of how many requests didn't get served because the server was busy. I know that being able to withstand the onslaught without server going down is an accomplishment and I am glad to know that ASP.NET is robust, but the ultimate goal is to serve resources so know how many didn't get served is important IMHO.