bigdata processing in the cloud – guest lecture - university of applied sciences rapperswil -...

98
© 2013 IBM Corporation 1 BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14 Romeo Kienzler IBM Innovation Center Source: http://res.sys-con.com/story/oct12/2398990/Cloud_BigData_468.jpg

Upload: romeo-kienzler

Post on 02-Dec-2014

249 views

Category:

Technology


0 download

DESCRIPTION

 

TRANSCRIPT

Page 1: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation1

BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

Romeo Kienzler

IBM Innovation Center

Source: http://res.sys-con.com/story/oct12/2398990/Cloud_BigData_468.jpg

Page 2: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation2

What is BIG data?

Page 3: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation3

What is BIG data?

Page 4: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation4

What is BIG data?

Big Data

Hadoop

Page 5: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation5

What is BIG data?

Business Intelligence

Data Warehouse

Page 6: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation6

Map-Reduce → Hadoop → BigInsights

Page 7: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation7

BigData UseCases● Google Index

● 40 X 10^9 = 40.000.000.000 => 40 billion pages indexed● Will break 100 PB barrier soon● Derived from MapReduce● now “caffeine” based on “percolator”

● Incremental vs. batch● In-Memory vs. disk

Page 8: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation8

BigData UseCases● CERN LHC

● 25 petabytes per year● Facebook

● Hive Datawarehouse● 300 PB, growing 600 TB / d● > 100 k servers

● Genomics● Enterprises

● Data center analytics (Logflies, OS/NW monitors, ...)● Predictive Maintenance, Cybersecurity

● Social Media Analytics● DWH offload● Call Detail Record (CDR) data preservation

http://www.balthasar-glaettli.ch/vorratsdaten/

Page 9: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation9

BigData Analytics

Page 10: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation10

BigData Analytics – Predictive Analytics

"sometimes it's not who has the best algorithm that wins; it's who has the most data."

(C) Google Inc.

The Unreasonable Effectiveness of Data¹

¹http://www.csee.wvu.edu/~gidoretto/courses/2011-fall-cp/reading/TheUnreasonable%20EffectivenessofData_IEEE_IS2009.pdf

No Sampling => Work with full dataset => No p-Value/z-Scores anymore

Page 11: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation11

Data Parallelism

Page 12: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation12

Aggregated Bandwith between CPU, Main Memory and Hard Drive

1 TB (at 10 GByte/s)

- 1 Node - 100 sec

- 10 Nodes - 10 sec

- 100 Nodes - 1 sec

- 1000 Nodes - 100 msec

Page 13: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation13

Fault Tolerance / Commodity Hardware

AMD Turion II Neo N40L (2x 1,5GHz / 2MB / 15W), 8 GB RAM,

3TB SEAGATE Barracuda 7200.14

< CHF 500

100 K => 200 X (2, 4, 3) => 400 Cores, 1,6 TB RAM, 200 TB HD

MTBF ~ 365 d > 1,5 d

Source: http://www.cloudcomputingpatterns.org/Watchdog

Page 14: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation14

Page 15: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation15

Page 16: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation16

HDFS – Hadoop File System

Page 17: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation17

Page 18: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation18

Page 19: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation19

Page 20: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation20

Page 21: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation21

Page 22: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation22

Page 23: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation23

Page 24: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation24

Page 25: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation25

Page 26: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation26

Page 27: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation27

Page 28: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation28

Page 29: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation29

Page 30: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation30

Page 31: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation31

Page 32: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation32

Page 33: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation33

Page 34: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation34

Page 35: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation35

Map-Reduce

Source: http://www.cloudcomputingpatterns.org/Map_Reduce

Page 36: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation36

Page 37: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation37

Page 38: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation38

Page 39: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation39

Page 40: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation40

Page 41: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation41

Page 42: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation42

Page 43: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation43

Page 44: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation44

Page 45: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation45

Page 46: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation46

Page 47: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation47

Page 48: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation48

Page 49: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation49

Page 50: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation50

Page 51: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation51

Page 52: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation52

Page 53: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation53

Page 54: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation54

Page 55: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation55

Page 56: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation56

Page 57: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation57

Page 58: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation58

Page 59: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation59

Page 60: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation60

Page 61: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation61

Page 62: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation62

Page 63: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation63

Page 64: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation64

Page 65: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation65

Page 66: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation66

Page 67: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation67

Page 68: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation68

Page 69: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation69

Page 70: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation70

Page 71: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation71

Page 72: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation72

Page 73: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation73

Page 74: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation74

Page 75: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation75

Page 76: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation76

Page 77: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation77

What role is the cloud playing here?

Page 78: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation78

“Elastic” Scale-Out

Source: http://www.cloudcomputingpatterns.org/Continuously_Changing_Workload

Page 79: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation79

“Elastic” Scale-Out

of

Page 80: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation80

“Elastic” Scale-Out

of

CPU Cores

Page 81: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation81

“Elastic” Scale-Out

of

CPU Cores Storage

Page 82: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation82

“Elastic” Scale-Out

of

CPU Cores Storage

Page 83: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation83

“Elastic” Scale-Out

of

CPU Cores Storage Memory

Page 84: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation84

“Elastic” Scale-Out

of

CPU Cores Storage Memory

Page 85: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation85

“Elastic” Scale-Out

linear

Source: http://www.cloudcomputingpatterns.org/Elastic_Platform

Page 86: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation86

“Elastic” Scale-Out

linear

Source: http://www.cloudcomputingpatterns.org/Elastic_Platform

Page 87: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation87

BigData Scale-Out

How do Databases Scale-Out?

Page 88: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation88

BigData Scale-Out

How do Databases Scale-Out?

Page 89: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation89

How do Databases Scale-Out?

Shared Disk Architectures

Page 90: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation90

How do Databases Scale-Out?

Shared Disk Architectures

Page 91: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation91

How do Databases Scale-Out?

Shared Nothing Architectures

Page 92: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation92

Born on the cloud Databases

Source: http://www.constructioncloudcomputing.com/wp-content/uploads/2010/10/dreamstime_7360880-480x300.jpg

Source: http://www.cloudcomputingpatterns.org/Execution_Environment

Page 93: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation93

Google AppEngine

Google App Engine is a Platform as a Service (PaaS) offering that lets you build and run applications on Google’s infrastructure. App Engine applications are easy to build, easy to maintain, and easy to scale as your traffic and data storage needs change. With App Engine, there are no servers for you to maintain. You simply upload your application and it’s ready to go.

Source: http://www.cloudcomputingpatterns.org/Platform_as_a_Service_%28PaaS%29

Page 94: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation94

Google AppEngine Database Services

Page 95: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation95

Page 96: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation96

IBM BlueMix

BlueMix is a Platform as a Service Cloud, based on Cloud Foundry, employing Enterprise grade services enriched with IBM Software and hosted at SOFTLAYER

Page 97: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation97

IBM BlueMix, a Cloudfoundry runtime

Linux VM

Linux VM

CodeRuntime

Framework+

Droplet

Linux VMContainer Container Container

SQL

Push

SSO

Services:

...

DropletDroplet

Page 98: BigData processing in the cloud – Guest Lecture - University of Applied Sciences Rapperswil - 29.4.14

© 2013 IBM Corporation98

● Summary

● BigData is born on the cloud

● Cloud facilitates resource provisioning, configuration and deployment

● Highly innovative area

● Technology

● UseCases

● Links

● http://en.wikipedia.org/wiki/MapReduce

● http://www.se-radio.net/2013/12/episode-199-michael-stonebraker/

● Sign up for the free BlueMix beta

● http://bluemix.net

● Come to the BlueMix Days

● http://bit.ly/1lsIY8J

● Use our software

● Biginsights: http://www.ibm.com/software/data/infosphere/biginsights/quick-start/