Google has already achieved the enviable marketing distinction of turning its name into a verb. There's probably not an Internet user in the world (even Steve Ballmer) who hasn't accessed it frequently for search and mapping. But its enormous popularity and global reach place an unintended burden on the search giant: When it goes down, the entire Web is shaken.
That's exactly what happened on Thursday, May 14, when Google suffered a major failure. A routing error sent traffic to servers in Asia, creating what Google called "a traffic jam." No kidding. According to the company, 14 percent of its users experienced slowdowns or outages. Many accounts put the number of those inconvenienced quite a bit higher. And we can't even guess at how many people were seriously put out by subsequent outages including the Google Gmail failure last month. But this isn't the day to beat on Google or fret about the implications Google outages have for cloud computing.
[ For more on Google's long-term cloud play, see "Google at 11: Taking the battle to Microsoft" and "Google set to take on collaboration giants." | Is our Internet future in danger? See InfoWorld's special report on whether the Net's infrastructure can handle projected demand. ]
What got my attention this week was a study to be formally presented on Oct. 19 of Internet usage by Arbor Networks, which found that just 100 ASNs (autonomous system numbers) out of about 35,000 account for some 60 percent of traffic on the public Internet. Put another way, out of the 40,000 routed sites in the Internet, 30 large companies now generate and consume a disproportionate 30 percent of all Internet traffic, according to the two-year study.
Not surprisingly, the biggest kahuna of all the big kahunas is Google, which accounts for about 6 percent of all Internet traffic globally. The other big guys include Level3, LimeLight, Akamai, and Microsoft, in that order.
Get the independent advice and expertise you need to support a virtual workforce.
The increase in Linux popularity has increased the frequency and sophistication of malware attacks. Read this 2 page white paper now to learn how you can protect your Linux environment with real-time protection that is certified by all major Linux vendors.
Download now »
Ensuring acceptable application delivery will become even more difficult over the next few years. As a result, IT organizations need to ensure that the approach that they take to resolving the current application delivery challenges can scale to support the emerging challenges. This handbook elaborates on the key tasks associated with planning, optimization, management and control and provides decision criteria to help IT organizations choose appropriate solutions.
Download now »
A common misconception is that mid-range storage requirements are dramatically different than that of a larger enterprise. Mid-range storage users may require less capacity, but they have similar functionality and management requirements. This ESG paper examines mid-range storage needs and reviews a new solution that adjusts size while retaining value, performance and functionality.
Download now »
Sign up to receive InfoWorld Resource Alerts
256 Recommendations
Mmmm, DANG!, I just do not have the time to respond properly to this. But very briefly, this whole piece is suffering from the Escher Syndrome. You seem to be having trouble distinguishing the image from the background, or the issue from the environment.
Yes, everything you say about Google, the big kahuna, is quite the case. But Google is not the issue. Google has, due to it's size, enormous impact on the issue, but it is background not image. Very early in it's history Google became the largest single consumer of transit bandwidth (consumer in the sense of being one end or the other of a transmission), but while their sheer bulk skews any view of the internet, their "failures" of note, (including all you mention), involve an unanticipated interaction with transit providers. No, I am not saying Google has no responsibility, but I am saying they are the background to your scale issue and the image, is the consolidation and stratification of the transport layer.
You almost looked squarely at the issue when you cataloged the high traffic players. Level3, LimeLight, and Akamai are transit players. To that list I would add another quarter to half dozen of the biggest transit players in the class with Vertela, as well as the big three Telcos, before I worried to much about "to big to fail".
The issue you see is in the math. IPV4 is based in a mathematical assumption that all nodes are treated similarly. IPV6 mitigates this to some degree, but V6 features that do that, are still optional. All the while, the transit players listed above are all trying diligently to become "to big to fail"...