The Evolution of
Data Architecture
Wei-Chiu Chuang
2017. 10 @ NCKU
1
Who’s Wei-Chiu?
Data Value Chain
AI
Machine
Learning
Data Science
Analytics
Big Data
Decision making
Insight
Automated
Decision making
Hype (?)
3
Data is the new Oil
https://www.economist.com/news/leaders/2172165
6-data-economy-demands-new-approach-antitrust-
rules-worlds-most-valuable-resource
4
Fastest way to
transmit 5MB of
data in 1956
6
Fast forward 60
years… transmit
100PB of data in 2016
Once upon a time, processors double in
speed every 18 months …
 The “Moore’s Law”
stopped 10 years ago.
 CPU, RAM and disk almost
stopped improving in
speed ever since.
7
Processor speed has been stagnant
 But data is being generated
at ever increasing speed.
 Hardware improvement
cannot keep up with data
generation.
 Multi-threaded systems,
distributed systems are the
must.
8
Distributed Systems are hard
Programmability
Scalability
Consistency
Availability
Partition Tolerance
Fault Tolerance
9
Big Data/Parallel Computing/Distributed
Sys.
D HPCBig DataCloud
Distributed Systems
10
Scale out
11
Modern Data Architecture
How do you:
 transmit
 collect
 store
 compute
Petabyte+ storage on
1000+ compute nodes?
12
Modern Data Center
DataCenter
ToR
Server1
Server10
ToR
Server1
Server10
ToR
Server1
Server10
ToR
Server1
Server10
Aggr Aggr Aggr
Core Core
Internet
AR AR
10Gbps
10Gbps
1Gbps
13
GFS
 Master – slave architecture
 Separation of control plane and
data plane
 Low cost, commodity hardware
 Failures are norm, rather
than exceptions
 Balance availability and network
partition tolerance
Control
messages
Data
messages
GFS
Master
GFS
chunkservers
/foo/bar
GFS
client
14
MapReduce
 A very simple yet powerful
distributed programming model
 Share-nothing architecture
 Programmability
 Data-locality:
 ship compute to data, rather
than shipping data to compute
 Fault tolerance:
 Intermediate state is stored in
storage.
 Failed tasks can be restarted
easily.
Split 0
Split 1
Split 2
worker
worker
worker
Input files Map phase
worker
worker
Intermediate
files
Reduce
phase
Output 0
Output files
Output 1
master
assign
map
assign
reduce
15
Hadoop
16
Hadoop
 GFS, MapReduce inspired Hadoop
 Initially developed by Yahoo!
 Released in 2006.
 Used by most large enterprises
 Hadoop 3.0 beta 1!
17
2006 2008 2009 2010 2011 2012 2013
Core Hadoop
(HDFS,
MapReduce)
HBase
ZooKeeper
Solr
Pig
Core Hadoop
Hive
Mahout
HBase
ZooKeeper
Solr
Pig
Core Hadoop
Sqoop
Avro
Hive
Mahout
HBase
ZooKeeper
Solr
Pig
Core Hadoop
Flume
Bigtop
Oozie
HCatalog
Hue
Sqoop
Avro
Hive
Mahout
HBase
ZooKeeper
Solr
Pig
YARN
Core Hadoop
Spark
Tez
Impala
Kafka
Drill
Flume
Bigtop
Oozie
HCatalog
Hue
Sqoop
Avro
Hive
Mahout
HBase
ZooKeeper
Solr
Pig
YARN
Core Hadoop
Parquet
Sentry
Spark
Tez
Impala
Kafka
Drill
Flume
Bigtop
Oozie
HCatalog
Hue
Sqoop
Avro
Hive
Mahout
HBase
ZooKeeper
Solr
Pig
YARN
Core Hadoop
2007
Solr
Pig
Core Hadoop
Knox
Flink
Parquet
Sentry
Spark
Tez
Impala
Kafka
Drill
Flume
Bigtop
Oozie
HCatalog
Hue
Sqoop
Avro
Hive
Mahout
HBase
ZooKeeper
Solr
Pig
YARN
Core Hadoop
2014 2015
Kudu
RecordService
Ibis
Falcon
Knox
Flink
Parquet
Sentry
Spark
Tez
Impala
Kafka
Drill
Flume
Bigtop
Oozie
HCatalog
Hue
Sqoop
Avro
Hive
Mahout
HBase
ZooKeeper
Solr
Pig
YARN
Core Hadoop
Evolution of the Hadoop Platform
 The stack is continually evolving and growing!
18
Mix and match
Resource Management
YARN Mesos Kubernetes
Storage
HDFS HBase Kudu S3 ADLS
Compute
MapReduce Hive Impala Spark Presto
Pig Drill Solr Storm
Ingest
Kafka
Flume
Beam
Samza
19
Open source in infra & platform
20
Why open source?
 It’s free ($$$)
 No vendor lock-in.
 Faster development and faster adoption.
 A new approach to foster collaboration.
 Open source software is becoming the standard.
21
Sell open source software, really?
 Water is free, but bottled water is not.
 Cloudera sells the “bottle”
 Cloudera’s Distribution of Hadoop.
 The integration of software.
 The support and services.
 The management software is
proprietary. The OSS is free of charge.
22
Market for open source software?
23
0
50
100
150
200
250
300
350
400
FY2015 FY2016 FY2017 FY2018 (f)
Revenue (million USD)
Hortonworks Cloudera MongoDB
Open Source Business Model
• MySQL
Dual licensing
• RedHat, Hortonworks
Support + services
• Java EE, Qt
Open core
• DataBricks, Amazon AWS, Microsoft Azure
Software as a Service
• Google Chrome, Android
Advertising-supported
• Cloudera, Confluent, MongoDB
Hybrid Open Source Software
24
Use Cases
25
“Big Data” finds many applications
across many industries
IT Healthcare Transportation Retail
Utilities Telecomm Public sector Manufactring
27
Applications and Use cases
 Realtime database for serving internet traffic
 Internet services (Facebook messenger), Twitter, Uber, Airbnb …
 Data analytics
 Assist in the development of new drugs by analyzing millions
of medical records
 Data science / Machine learning
 Fraud detection
 Anti-money laundry
 Cybersecurity
28
Fraud Detection System using Hadoop
The Cloudera Platform for IoT – Data Mgmt. Value Chain
Data Sources Data Ingest Data Storage & Processing
Serving, Analytics &
Machine Learning
ENTERPRISE DATA HUB
Apache Kafka
Stream or batch ingestion of IoT data
Apache Sqoop
Ingestion of data from relational sources
Apache Hadoop
Storage (HDFS) & deep batch processing
Apache Kudu
Storage & serving for fast changing data
Apache HBase
NoSQL data store for real time
applications
Apache Impala
MPP SQL for fast analytics
Cloudera Search
Real time searchConnected Things/ Data
Sources
Other Data Sources Security, Scalability & Easy Management
Deployment Flexibility:
Datacenter Cloud
Apache Spark
Stream & iterative processing, ML
IoT Use Case 1:
Predictive Maintenance
Predictive Maintenance on Thousands of
Industrial Machinery in Real- Time
Challenge:
• Collect and analyze data from
thousands of diverse manufacturing
systems in real-time
Solution:
• iTrak application using Cloudera in
the Cloud to monitor the performance
of individual manufacturing systems
in real-time
• Predictive Maintenance - Proactively
identifying & fixing issues before
they break
MANUFACTURING
» INDUSTRIAL IoT
» PREDICTIVE MAINTENANCE
» IMPROVED EFFICIENCIES
Industrial IoT – Predictive Maintenance
DATA-DRIVEN
PROCESS
CASE STUDY
DATA-DRIVEN
PRODUCTS
Use Case 2:
Connected Vehicles
Using Predictive Maintenance to Improve
Performance and Reduce Fleet Downtime
Challenge:
• Monitor the health of 180,000+ trucks
in real-time in order to minimize
downtime
Solution:
• OnCommand Connection collecting
telematics and geolocation data across
thousands of trucks
• Identify and correct engine problems
early, and increase fleet uptime
• Reduced maintenance costs to $.03
per mile from $.12-$.15 per mile
Connected Vehicles & Telematics
DATA-DRIVEN
PROCESS
CASE STUDY
DATA-DRIVEN
PRODUCTS
TRANSPORTATION
» PREDICTIVE MAINTENANCE
» TELEMETRY
» LOWER TCO
Use Case 3:
Smart Cities & Smart Infrastructure
Enabling the State of Kentucky manage
snow and ice events in real time
Challenge:
• Kentucky Transportation Cabinet (KYTC)
oversees the state’s transportation system,
which includes 27,000 miles of highways, 230
airports and heliports, and more than three
million drivers.
• Needed more efficient approach to inclement
weather road management
Solution:
• KYTC has built a real-time weather response
system that incorporates real-time data from
Waze, HERE, ESRI’s GeoEvent processor, and
Automatic Vehicle Locations (providing
sensor data from salt trucks).
• KYTC aggregates 15-20 million records every
day and process more than a million records
per second.
Data Driven Dept. of Transportation
Source: http://www.routefifty.com/2016/09/data-drives-government/131821/
2016 Data Impact Award Winner
State of Kentucky Department of
Transportation
Use Case 4:
Connected Healthcare
Improve Parkinson's Disease
Monitoring and Treatment through IoT
Challenge:
• Collect and analyze data from
wearables (more than 300 readings
per second) from thousands of
patients in real-time
Solution:
• Cloudera on Intel architecture to
detect patterns in patient data
streaming from wearables
• Continuously monitor the patients
and symptoms to understand the
progression of the disease
objectively
HEALTHCARE
» WEARABLES
» PREDICTIVE ANALYTICS
» IMPROVED CARE
Connected Healthcare
DATA-DRIVEN
PROCESS
CASE STUDY
DATA-DRIVEN
PRODUCTS
Building a Holistic Picture of the US
Securities Market From 50 Billion Daily
Events
• Saving $10-20M in operational
efficiencies annually
• 90-minute queries run in 10 seconds
• Supporting future market growth and a
dynamic regulatory environment.
CUSTOMER 360
Using Big Data to Help Consumers Save
Hundreds of Millions in Utility Bills
• Relevant insight into household energy
use improves energy consciousness
• 2.7+ TWH (terawatt hours) saved to
date
• Motivated consumers to save enough
energy to power every household in Salt
Lake City and St. Louis for a year
CUSTOMER 360
ENERGY & UTILITIES
» PRODUCT INNOVATION
» SERVICE IMPROVEMENT
» IOT
Saving Lives by Detecting Sepsis Early
Enough for Successful Treatment
• Builds a more complete picture of
patients, conditions, and trends
• Has saved 100’s of lives already
• Reduces hospital readmissions
• 2PB+ in multi-tenant environment
supporting 100s of clients
• Secure yet explorable
HEALTHCARE
» 360° CUSTOMER VIEW
» PREDICTIVE ANALYTICS
» IMPROVED SERVICE
Improving Pediatric Care and Outcomes
• Quantifying effect of ambient noise on
children’s vital signs
• Identifying cancerous genome variants
in 20 minutes (vs. days before)
• Performing fewer CT scans and higher
quality surgeries
CUSTOMER 360
HEALTHCARE
» MACHINE LEARNING
» IOT
» 360o CUSTOMER VIEW
Government Revenue Service
Increasing Customer Convenience
• Provides view of the complete taxpayer
journey
• Creates ability to pre-populate tax
returns for increased ease of use
• Supports move to near-real-time
oversight of operations and faster
response
CUSTOMER 360
GOVERNMENT
» SERVICE IMPROVEMENT
» PROCESS IMPROVEMENT
» 360° CUSTOMER VIEW
Driving Growth and Innovation
• Combines 80+ years’ data spanning all
business units and 50 states
• Expedites holistic analysis and reports
by 500X
• Enables more accurate and detailed
predictive models to customize offers,
optimizing pricing, and minimize risk
CUSTOMER 360
INSURANCE
» 360° CUSTOMER VIEW
» FRAUD DETECTION
» PREDICTIVE ANALYTICS
Re-Platformed 1,600 Operational
Databases & Systems onto a Cloudera EDH
• Business & consumer data was spread
over a dozen different customer
databases
• One daily ETL job (processing 1 billion
customer records) used to take 24 hours
• Increased data velocity by 15x
(5 times the data in 1/3 of the time)
Now completes in 1 ½ hours
• BT now has access to the most up-to-
date and centralized data for all their
customers
CUSTOMER
360
TELECOMMUNICATIONS
» IMPROVED SERVICE
» PROCESS IMPROVEMENT
» IT COST REDUCTION
Future
48
Future
 Hardware evolution:
 Cloud
 40Gbps, 100Gbps networks
 GPU, TPU
 Flash disk
 Application-driven:
 Machine learning, deep learning
 Realtime data stream processing (IoT)
49
Future
How to scale by an order of
magnitude in 5 years?
We are here today
In 10 years?
50
台灣資料工程協會
Click to enter confidentiality information
台灣人參與Apache
Click to enter confidentiality information
葉祐欣 謝良奇、蔡東邦 陳恩平
戴資力 莊偉赳 蔡嘉平
Apache Contributor 育才賽
Click to enter confidentiality information
Takeaway
If you only remember 3 things from this talk:
1.Data is the new Oil
2.Open source is the standard
3.Think big! Remember GFS:
failures are the norm rather
than the exception!
54
Thank you
jojochuang@gmail.com / weichiu@apache.org / weichiu@cloudera.com
55

The Evolution of Data Architecture

  • 1.
    The Evolution of DataArchitecture Wei-Chiu Chuang 2017. 10 @ NCKU 1
  • 2.
  • 3.
    Data Value Chain AI Machine Learning DataScience Analytics Big Data Decision making Insight Automated Decision making Hype (?) 3
  • 4.
    Data is thenew Oil https://www.economist.com/news/leaders/2172165 6-data-economy-demands-new-approach-antitrust- rules-worlds-most-valuable-resource 4
  • 5.
    Fastest way to transmit5MB of data in 1956
  • 6.
    6 Fast forward 60 years…transmit 100PB of data in 2016
  • 7.
    Once upon atime, processors double in speed every 18 months …  The “Moore’s Law” stopped 10 years ago.  CPU, RAM and disk almost stopped improving in speed ever since. 7
  • 8.
    Processor speed hasbeen stagnant  But data is being generated at ever increasing speed.  Hardware improvement cannot keep up with data generation.  Multi-threaded systems, distributed systems are the must. 8
  • 9.
    Distributed Systems arehard Programmability Scalability Consistency Availability Partition Tolerance Fault Tolerance 9
  • 10.
    Big Data/Parallel Computing/Distributed Sys. DHPCBig DataCloud Distributed Systems 10
  • 11.
  • 12.
    Modern Data Architecture Howdo you:  transmit  collect  store  compute Petabyte+ storage on 1000+ compute nodes? 12
  • 13.
  • 14.
    GFS  Master –slave architecture  Separation of control plane and data plane  Low cost, commodity hardware  Failures are norm, rather than exceptions  Balance availability and network partition tolerance Control messages Data messages GFS Master GFS chunkservers /foo/bar GFS client 14
  • 15.
    MapReduce  A verysimple yet powerful distributed programming model  Share-nothing architecture  Programmability  Data-locality:  ship compute to data, rather than shipping data to compute  Fault tolerance:  Intermediate state is stored in storage.  Failed tasks can be restarted easily. Split 0 Split 1 Split 2 worker worker worker Input files Map phase worker worker Intermediate files Reduce phase Output 0 Output files Output 1 master assign map assign reduce 15
  • 16.
  • 17.
    Hadoop  GFS, MapReduceinspired Hadoop  Initially developed by Yahoo!  Released in 2006.  Used by most large enterprises  Hadoop 3.0 beta 1! 17
  • 18.
    2006 2008 20092010 2011 2012 2013 Core Hadoop (HDFS, MapReduce) HBase ZooKeeper Solr Pig Core Hadoop Hive Mahout HBase ZooKeeper Solr Pig Core Hadoop Sqoop Avro Hive Mahout HBase ZooKeeper Solr Pig Core Hadoop Flume Bigtop Oozie HCatalog Hue Sqoop Avro Hive Mahout HBase ZooKeeper Solr Pig YARN Core Hadoop Spark Tez Impala Kafka Drill Flume Bigtop Oozie HCatalog Hue Sqoop Avro Hive Mahout HBase ZooKeeper Solr Pig YARN Core Hadoop Parquet Sentry Spark Tez Impala Kafka Drill Flume Bigtop Oozie HCatalog Hue Sqoop Avro Hive Mahout HBase ZooKeeper Solr Pig YARN Core Hadoop 2007 Solr Pig Core Hadoop Knox Flink Parquet Sentry Spark Tez Impala Kafka Drill Flume Bigtop Oozie HCatalog Hue Sqoop Avro Hive Mahout HBase ZooKeeper Solr Pig YARN Core Hadoop 2014 2015 Kudu RecordService Ibis Falcon Knox Flink Parquet Sentry Spark Tez Impala Kafka Drill Flume Bigtop Oozie HCatalog Hue Sqoop Avro Hive Mahout HBase ZooKeeper Solr Pig YARN Core Hadoop Evolution of the Hadoop Platform  The stack is continually evolving and growing! 18
  • 19.
    Mix and match ResourceManagement YARN Mesos Kubernetes Storage HDFS HBase Kudu S3 ADLS Compute MapReduce Hive Impala Spark Presto Pig Drill Solr Storm Ingest Kafka Flume Beam Samza 19
  • 20.
    Open source ininfra & platform 20
  • 21.
    Why open source? It’s free ($$$)  No vendor lock-in.  Faster development and faster adoption.  A new approach to foster collaboration.  Open source software is becoming the standard. 21
  • 22.
    Sell open sourcesoftware, really?  Water is free, but bottled water is not.  Cloudera sells the “bottle”  Cloudera’s Distribution of Hadoop.  The integration of software.  The support and services.  The management software is proprietary. The OSS is free of charge. 22
  • 23.
    Market for opensource software? 23 0 50 100 150 200 250 300 350 400 FY2015 FY2016 FY2017 FY2018 (f) Revenue (million USD) Hortonworks Cloudera MongoDB
  • 24.
    Open Source BusinessModel • MySQL Dual licensing • RedHat, Hortonworks Support + services • Java EE, Qt Open core • DataBricks, Amazon AWS, Microsoft Azure Software as a Service • Google Chrome, Android Advertising-supported • Cloudera, Confluent, MongoDB Hybrid Open Source Software 24
  • 25.
  • 26.
    “Big Data” findsmany applications across many industries IT Healthcare Transportation Retail Utilities Telecomm Public sector Manufactring 27
  • 27.
    Applications and Usecases  Realtime database for serving internet traffic  Internet services (Facebook messenger), Twitter, Uber, Airbnb …  Data analytics  Assist in the development of new drugs by analyzing millions of medical records  Data science / Machine learning  Fraud detection  Anti-money laundry  Cybersecurity 28
  • 28.
  • 29.
    The Cloudera Platformfor IoT – Data Mgmt. Value Chain Data Sources Data Ingest Data Storage & Processing Serving, Analytics & Machine Learning ENTERPRISE DATA HUB Apache Kafka Stream or batch ingestion of IoT data Apache Sqoop Ingestion of data from relational sources Apache Hadoop Storage (HDFS) & deep batch processing Apache Kudu Storage & serving for fast changing data Apache HBase NoSQL data store for real time applications Apache Impala MPP SQL for fast analytics Cloudera Search Real time searchConnected Things/ Data Sources Other Data Sources Security, Scalability & Easy Management Deployment Flexibility: Datacenter Cloud Apache Spark Stream & iterative processing, ML
  • 30.
    IoT Use Case1: Predictive Maintenance
  • 31.
    Predictive Maintenance onThousands of Industrial Machinery in Real- Time Challenge: • Collect and analyze data from thousands of diverse manufacturing systems in real-time Solution: • iTrak application using Cloudera in the Cloud to monitor the performance of individual manufacturing systems in real-time • Predictive Maintenance - Proactively identifying & fixing issues before they break MANUFACTURING » INDUSTRIAL IoT » PREDICTIVE MAINTENANCE » IMPROVED EFFICIENCIES Industrial IoT – Predictive Maintenance DATA-DRIVEN PROCESS CASE STUDY DATA-DRIVEN PRODUCTS
  • 32.
  • 33.
    Using Predictive Maintenanceto Improve Performance and Reduce Fleet Downtime Challenge: • Monitor the health of 180,000+ trucks in real-time in order to minimize downtime Solution: • OnCommand Connection collecting telematics and geolocation data across thousands of trucks • Identify and correct engine problems early, and increase fleet uptime • Reduced maintenance costs to $.03 per mile from $.12-$.15 per mile Connected Vehicles & Telematics DATA-DRIVEN PROCESS CASE STUDY DATA-DRIVEN PRODUCTS TRANSPORTATION » PREDICTIVE MAINTENANCE » TELEMETRY » LOWER TCO
  • 34.
    Use Case 3: SmartCities & Smart Infrastructure
  • 35.
    Enabling the Stateof Kentucky manage snow and ice events in real time Challenge: • Kentucky Transportation Cabinet (KYTC) oversees the state’s transportation system, which includes 27,000 miles of highways, 230 airports and heliports, and more than three million drivers. • Needed more efficient approach to inclement weather road management Solution: • KYTC has built a real-time weather response system that incorporates real-time data from Waze, HERE, ESRI’s GeoEvent processor, and Automatic Vehicle Locations (providing sensor data from salt trucks). • KYTC aggregates 15-20 million records every day and process more than a million records per second. Data Driven Dept. of Transportation Source: http://www.routefifty.com/2016/09/data-drives-government/131821/ 2016 Data Impact Award Winner State of Kentucky Department of Transportation
  • 36.
  • 37.
    Improve Parkinson's Disease Monitoringand Treatment through IoT Challenge: • Collect and analyze data from wearables (more than 300 readings per second) from thousands of patients in real-time Solution: • Cloudera on Intel architecture to detect patterns in patient data streaming from wearables • Continuously monitor the patients and symptoms to understand the progression of the disease objectively HEALTHCARE » WEARABLES » PREDICTIVE ANALYTICS » IMPROVED CARE Connected Healthcare DATA-DRIVEN PROCESS CASE STUDY DATA-DRIVEN PRODUCTS
  • 38.
    Building a HolisticPicture of the US Securities Market From 50 Billion Daily Events • Saving $10-20M in operational efficiencies annually • 90-minute queries run in 10 seconds • Supporting future market growth and a dynamic regulatory environment. CUSTOMER 360
  • 39.
    Using Big Datato Help Consumers Save Hundreds of Millions in Utility Bills • Relevant insight into household energy use improves energy consciousness • 2.7+ TWH (terawatt hours) saved to date • Motivated consumers to save enough energy to power every household in Salt Lake City and St. Louis for a year CUSTOMER 360 ENERGY & UTILITIES » PRODUCT INNOVATION » SERVICE IMPROVEMENT » IOT
  • 40.
    Saving Lives byDetecting Sepsis Early Enough for Successful Treatment • Builds a more complete picture of patients, conditions, and trends • Has saved 100’s of lives already • Reduces hospital readmissions • 2PB+ in multi-tenant environment supporting 100s of clients • Secure yet explorable HEALTHCARE » 360° CUSTOMER VIEW » PREDICTIVE ANALYTICS » IMPROVED SERVICE
  • 41.
    Improving Pediatric Careand Outcomes • Quantifying effect of ambient noise on children’s vital signs • Identifying cancerous genome variants in 20 minutes (vs. days before) • Performing fewer CT scans and higher quality surgeries CUSTOMER 360 HEALTHCARE » MACHINE LEARNING » IOT » 360o CUSTOMER VIEW
  • 42.
    Government Revenue Service IncreasingCustomer Convenience • Provides view of the complete taxpayer journey • Creates ability to pre-populate tax returns for increased ease of use • Supports move to near-real-time oversight of operations and faster response CUSTOMER 360 GOVERNMENT » SERVICE IMPROVEMENT » PROCESS IMPROVEMENT » 360° CUSTOMER VIEW
  • 43.
    Driving Growth andInnovation • Combines 80+ years’ data spanning all business units and 50 states • Expedites holistic analysis and reports by 500X • Enables more accurate and detailed predictive models to customize offers, optimizing pricing, and minimize risk CUSTOMER 360 INSURANCE » 360° CUSTOMER VIEW » FRAUD DETECTION » PREDICTIVE ANALYTICS
  • 44.
    Re-Platformed 1,600 Operational Databases& Systems onto a Cloudera EDH • Business & consumer data was spread over a dozen different customer databases • One daily ETL job (processing 1 billion customer records) used to take 24 hours • Increased data velocity by 15x (5 times the data in 1/3 of the time) Now completes in 1 ½ hours • BT now has access to the most up-to- date and centralized data for all their customers CUSTOMER 360 TELECOMMUNICATIONS » IMPROVED SERVICE » PROCESS IMPROVEMENT » IT COST REDUCTION
  • 45.
  • 46.
    Future  Hardware evolution: Cloud  40Gbps, 100Gbps networks  GPU, TPU  Flash disk  Application-driven:  Machine learning, deep learning  Realtime data stream processing (IoT) 49
  • 47.
    Future How to scaleby an order of magnitude in 5 years? We are here today In 10 years? 50
  • 48.
    台灣資料工程協會 Click to enterconfidentiality information
  • 49.
    台灣人參與Apache Click to enterconfidentiality information 葉祐欣 謝良奇、蔡東邦 陳恩平 戴資力 莊偉赳 蔡嘉平
  • 50.
    Apache Contributor 育才賽 Clickto enter confidentiality information
  • 51.
    Takeaway If you onlyremember 3 things from this talk: 1.Data is the new Oil 2.Open source is the standard 3.Think big! Remember GFS: failures are the norm rather than the exception! 54
  • 52.
    Thank you jojochuang@gmail.com /weichiu@apache.org / weichiu@cloudera.com 55