Web infrastructure is a hot topic today, after Amazon Web Services experienced an outage over the weekend, and after Facebook released some interesting details about its Hadoop cluster on Friday. Even LinkedIn is making headlines by expanding into a new Los Angeles data center. There’s also an interesting take on the link between VCs and cloud computing, which is a good complement to my story from this morning about CloudSwitch.
Amazon’s WikiLeaks Response Threatens Cloud Computing (From the Wall Street Journal) This is an easy viewpoint to take if you’re a competitive cloud provider operating largely in Europe. I’m as idealistic as the next guy, but this is business, and there are practical matters involving the law and shareholders.
Amazon: Hardware Failures Caused Outage (From Data Center Knowledge) As if cloud computing, and Amazon, particularly, needed any more bad press after the Wikileaks situation. AWS is probably more reliable than most company’s IT operations, but most just see the word “outage.”
Looking At the Code Behind Our Three Uses of Apache Hadoop (From Facebook Engineering’s Notes) Very interesting post from Facebook, which is running some innovative Hadoop tools and might have the world’s largest Hadoop cluster — 30PB.
Cloud Computing Is Growing Up, and Venture Capitalists Are Feeding It (From DailyFinance) And, as the author notes re: Heroku, it doesn’t take a lot of funding to result in a big sale anymore. IaaS lets cloud startups build their services on the cheap and most of the resources toward building products.
LinkedIn Moves to Los Angeles Data Center (From InformationWeek) Considering that LinkedIn is, essentially, Facebook for professionals, it’s not surprising that it needs to expand its web infrastructure to meet demand. Don’t hear much about its operations, though.
For more cloud-related news analysis and research, visit GigaOM Pro.