How Amazon saved Zynga’s butt – and why Zynga built a cloud of its own – Ars Technica, May 2012
After the game’s 2009 release, 10 million users were hitting FarmVille servers within six weeks, and 25 million within five months. “We couldn’t get power fast enough. We couldn’t get servers fast enough. We just couldn’t scale our infrastructure to match the needs of FarmVille,” said Allan Leinwand, Zynga’s CTO of infrastructure.
Zynga upended its whole IT model, shifting most of its infrastructure to the Amazon Elastic Compute Cloud. But eventually, Zynga “realized we could actually do it on our own and we could scale it in a way that worked better for our business.” Now, 80 percent of Zynga usage is accommodated by Zynga-built data centers, and the other 20 percent is fueled by Amazon. Zynga doesn’t plan to ditch Amazon entirely, as it’s still a great “shock absorber” for sudden increases in interest.
iCloud’s first six months: The developers weigh in -Mac Stories, Apr 2012
“It is a non-trivial undertaking, when you are dealing with files, folders, and conflicts”
When it works correctly, iCloud support in third-party iOS and Mac apps has enabled developers to build natural, more connected experiences that leverage the inner strengths of the cloud to let users forego the process of copying files, or set an app’s preferences every time it runs on a new device.
…For better or worse, the past six months have showed that iCloud — as a platform for Apple to build the future of iOS and a syncing technology developers can use in their apps — paves the way for a more intuitive, connected ecosystem of apps that “just work” with a user’s content always available, and always up-to-date.
Introducing Amazon Cloud Services – Werner Vogels, Amazon CTO, Apr 2012
Amazon CloudSearch provides a fully-featured search engine that is easy to manage and scale. It offers full-text search with features like faceting and user-defined rank functions… There are three features that make it easy to configure and customize the search results to meet exactly the needs of the application.
Filtering: Conceptually, this is using a match in a document field to restrict the match set. For example, if documents have a “color” field, you can filter the matches for the color “red”.
[Customised] Ranking: Search has at least two major phases: matching and ranking. The query specifies which documents match, generating a match set. After that, scores are computed (or direct sort criterion is applied) for each of the matching documents to rank them best to worst.
Faceting: Faceting allows you to categorize your search results into refinements on which the user can further search. For example, a user might search for ‘umbrellas’, and facets allow you to group the results by price, such as $0-$10, $10-$20, $20-$40, etc.
“It just works” – announcing iCloud – MG Siegler, TechCrunsh, Jun 2011
Apple’s belief is clearly that users will not and should not care how the cloud actually works… Apple has been going out of their way to avoid using the word “syncing” with regard to iCloud… Apple has rethought and rewritten their apps — including their desktop apps — from the ground up to be woven with iCloud fabric that a user won’t see.
Microsoft BPOS cloud customers hit by multi-day email outage – Mary-Jo Foley, May 2011
Microsoft officials warned customers of its BPOS bundle of hosted Exchange, SharePoint and Lync on Monday, May 10, that an upgrade of Exchange Online was slated to begin on May 12. Microsoft didn’t tell users to expect any downtime as a result of the upgrade. But it seems something went wrong before May 12’s upgrade ever began.
Netflix, Amazon and the Chaos Monkey – Apr 11 – Coding Horror
One of the first systems our engineers built in AWS is called the Chaos Monkey. The Chaos Monkey’s job is to randomly kill instances and services within our architecture. If we aren’t constantly testing our ability to succeed despite failure, then it isn’t likely to work when it matters most – in the event of an unexpected outage.
Google: A study in scalability -Jeff Dean talk at Stanford, Nov 2010
From 1999 to 2010, docs and queries up 1000x, query latency decreased 5x
From 2007 to 2010, data doubled but no. machines remained constant
Don’t design to scale infinitely – consider 5X – 50X growth. But > 100X requires redesign
In 2004, did a rethink and refreshed systems infrastructure from scratch
Facebook to build it’s own datacenters – Data Center Knowledge, Jan 2010
The new design foregoes traditional uninterruptible power supply (UPS) and power distribution units (PDUs) and adds a 12 volt battery to each server power supply. This approach was pioneered by Google, the company cited this design as a key factor in the exceptional energy efficiency data for its data centers
A peak behind the scenes at Hotmail – Windows team blog, Dec 2009
Host 1.3 billion inboxes, 350 million active users, 3 billion messages a day filtering out over 1 billion spam messages, storage growing at over 2 petabytes per month. Service is deployed in clusters to scale.
More links
- Microsoft Cloud Computing events (online web casts)
- Microsoft Office 365 materials (Partner site)
- Google App Engine – write your own apps on Google’s infrastructure
- Microsoft Private Cloud Security presentation – Channel 9, May 2011
- Office 365 now FISMA-certified – ZDNet, May 2012
