Your SlideShare is downloading. ×
0
Strata 2012 Million Monkeys
Strata 2012 Million Monkeys
Strata 2012 Million Monkeys
Strata 2012 Million Monkeys
Strata 2012 Million Monkeys
Strata 2012 Million Monkeys
Strata 2012 Million Monkeys
Strata 2012 Million Monkeys
Strata 2012 Million Monkeys
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×
Saving this for later? Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime – even offline.
Text the download link to your phone
Standard text messaging rates apply

Strata 2012 Million Monkeys

192

Published on

Published in: Technology
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
192
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
4
Comments
0
Likes
0
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide
  • Interesting statistical question. Thought about since Aristotle.Randomness+Resouces+Time=AnythingPossibleNo real monkeys – need virtual monkeys
  • Lucky monkeyThe monkey wears a lot of hats. He generates and then compares.Every work of Shakespeare created. First was A Lover’s Complaint and last was Taming of the ShrewVisualization to find your favorite line from Shakespeare
  • Shakespeare lazy. Heavily influenced English Literature.Big Data isn’t always a huge file. It can be high computation.
  • Creating Shakespeare not a business. Don’t have Shakespeare in your data.If you look hard enough you will find itHumans are not randomYou want to be looking for what’s actually there. Check your assumptionsOperate with scientific method. Form a hypothesis. Test hypothesis against data.Offer what customers are looking for. Not what you think or favorite or new product. Only what your data shows.
  • This is not a map of MT and ID1 to 20 node testingKeep efficiency up RDBMS efficiency in gutter
  • Engineers not spending time coding to scale. Busy adding new features.No code changes for scaling. Took 1.5 months on one computer and 3.5 days on 20 nodesSpending on new computers gives a consistent, linear increase. Compare spending on RDBMS and Hadoop.
  • We like to ask bigger questions.I asked if Shakespeare could be randomly recreated by a bunch of virtual monkeys? The answer is yes.
  • Transcript

    • 1. Given Enough MonkeysSome Thoughts on RandomnessJesse Anderson | CLOUDERA, INSTRUCTOR
    • 2. Infinite Monkey Theorem2
    • 3. Million Monkeys Algorithm Randomly generate a 9 character group TOBEORNOT Does it exist in Shakespeare? To be, or not to be- that is the question3
    • 4. Exponential Growth (aka Big Data) Odds of finding a group Contiguous Combinations of characters is 1 in 26 Characters raised to the power of the number of 8 208,827,064,576 contiguous characters 9 5,429,503,678,976 10 141,167,095,653,3764
    • 5. Data Bias?5
    • 6. Hadoop Scalability Percent of Linear Scalability 100 80 Percent 60 RDBMS Hadoop 40 20 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Nodes RDBMS = Relational Database6
    • 7. Business Value of Scalability Scaling does not require Adding more computers massive re-engineering to cluster gets a and complete rewrites of predictable increase in code computational power and storage SAVE SAVE7
    • 8. Going Viral (and taking over the world) Covered internationally 26,000 unique in BBC, Wall Street visits from 119 Journal, Wired and countries in Slashdot one day8
    • 9. @jessetanderson

    ×