Hey, it's HighScalability time:

Who wants in on the over? We are not alone if the probability a habitable zone planet develops a technological species is larger than 10^-24.

• 100,000+: bare metal servers run by Twitter; 10 billion: Snapchat videos delivered daily; $2.57 billion: AWS fourth quarter revenues; 40 light years: potentially habitable planets; 1700: seed banks around the world; 560x: throughput after SSD optimization; 12: data science algorithms; $2.8 billion: new value of Pivotal’s Cloud Foundry;  


• Quotable Quotes:
  • @skap5: Pied Piper's product is its stock and anything that makes its price go up! #SiliconValley
  • Seth Godin: It pays to have big dreams but low overhead. 
  • Craig Venter~ Our knowledge of the genome hasn't changed a lot since 2003, but it's about to start changing rapidly. One of the key things for understanding the genome is to get very large numbers of genomes so we can understand out of the 6.2 billion or so letters of genetic code the less than 3% that we have different amongst the entire human population. We need very large data sets to understand the differences and significances. That's where the cost and speed of sequencing has had such an immediate impact. 
  • @elonmusk: Rocket reentry is a lot faster and hotter than last time, so odds of making it are maybe even, but we should learn a lot either way
  • @EconBizFin: The space race was once between capitalism and communism. Now it's individual capitalists
  • Grit: substitute nuance for novelty. Rather than constantly moving on to a new thrill try to find another dimension of the thing you are already doing to make it more thrilling. 
  • David Rosenthal: Overall the message is: Storage Will Be Much Less Free Than It Used To Be
  • StorageMojo: The losers are the systems that make customers pay for features they no longer need. Winners will successfully blend ease of use with performance and availability – at a competitive price.
  • Tim Harford: These distractions were actually grists to their creative mill. They were able to think outside the box because their box was full of holes.
  • Juan Enriquez~ Plastics was the wrong advice to Dustin Hoffman in The Graduate. The word should have been silicon. In 2015 the word is lifecode, the various means we have to program life.
  • Benjamin Treynor Sloss: If you've ever wondered about how run reliable services, this beautifully written intro from the SRE book is the best 5-minute guide on the topic.
  • cs702: Without AWS, Amazon would have reported losses! 
  • Kode Vicious: A single cache miss is more expensive than many instructions.
  • @aminggs: “your database… unlikely to provide serializability, your multi-core processor… unlikely to provide linearizability” 
  • @mrogati: A decade in academia taught me a bunch of sophisticated algorithms; a decade in industry taught me when not to use them.
  • @mjpt777: Hardware tries so hard to make software fast; software tries so hard to make hardware slow.
  • @jyarow: Echo sold 3 million units. Gets stories that it’s next great business for Amazon. Apple Watch sold 12 million units, gets panned as a flop.
  • @balajis: 5/ At that time, the highest truth comes not from faith in god or trust in the state, but from the ability to check the math of the network.
  • Benedict Evans: The smartphone install base does have a lot of room to grow, but that's a function of replacement at close to existing volumes, and even that will be largely done in a few more years. Hence: smartphone sales growth is slowing down. 
  • @giupan: Colocated teams where Devs are sitting together with Product and UX outperform distributed teams. Don't split up skills @cagan #craftconf
  • Mathias Bynens: To me, this stuff is extremely interesting on a technical level. It’s also a little scary, however, to realize that malicious actors can use these techniques to invade your privacy while you’re browsing the web, without you ever knowing.
  • Le Corbusier: yes, the Parthenon is perhaps the most beautiful instance, the perfect example of a particular standard of architecture. The Parthenon may have achieved the platonic ideal of the standard of architecture we’ve previously established. But there are many possible standards to acknowledge, each dependent on need and use, and standards are established by experiment.
  • PaulHoule: Atom chips have always been crippled to keep them from cannibalizing more expensive chips. Skylake is a fine tablet chip, in fact, that's really what Skylake is good for. They are probably producing them in high enough numbers now that they can give up on Atom
  • Chau Tu: CyArk wants to preserve our world’s important cultural heritage sites before they turn to dust...with new imaging technologies to steadily build a digital archive of the past, for the future. 
  • Neill Turner: Over time i think OpenStack will be a niche product for large corporates that don’t want to use public clouds. For everyone else they with be doing hybrid IT – that is extending their existing IT infrastructure into the Public Cloud. When they see what is left to run to run outside public clouds then they can see where to take that portion of the workloads.


• What if going to Mars is how we fix our economy? Trump: Before going to Mars, America needs to fix its economy. A problem can't be solved at the same level it was created. Someone smart said that once. Isn't expanding the economy in to space the only way we'll be able to generate the constant growth a modern economy so desperately devowers? Walls don't lift boats.


• Is it dystopian to hire real meat people to train your AI? Interesting question posed by John Robb in a tweet: "they were there not to work, but to serve as training modules for Facebook’s algorithm" Journos at Fbook


• Peter Bailis offers in a heartfelt visionary article four pieces of advice to get the database community out of its identity crisis: 1) Kill the reference architecture and rethink our conception of “database.” 2) Solve new, emerging, real problems outside traditional relational database systems. 3) Use data-intensive tools, both the tools that you’re building and the tools that others have built. 4) Do bold, weird, and hard projects and actually follow through. Examples in action: Peter Alvaro’s work on Molly and Lineage-Driven Fault Injection; Chris Ré’s work on DeepDive; A recent project I wish the database community had done is TensorFlow at Google. 

Don't miss all that the Internet has to say on Scalability, click below and become eventually consistent with all scalability knowledge (which means this post has many more items to read so please keep on reading)...