Você está na página 1de 37

In-Memory Processing for High Performance Analytics

We will begin momentarily

Ask Questions
There is a <Q&A> module in the upper right corner of your screen.
Ask questions as they come to mind throughout the presentation. We
will address as many as possible as time allows.
There are also additional icons to explore
The Resource Links icon will allow you to download todays presentation, as
well as other relevant materials
Find more info about the todays Speakers.
And there is also a Help button for assistance with any technical difficulties
you might experience. Please hit F5 to refresh your console as a first step.

2015 Teradata

Todays Speakers
Noel Yuhanna
Principal Analyst serving Enterprise Architecture Professionals
Forrester Research
Noel covers big data, data warehouses, Hadoop, in-memory, databases, NoSQL, data integration, data
virtualization, data archiving, cloud, ETL, information fabric, big data integration, data management, data tools,
and data security for Enterprise Architecture Professionals. His current focus is on Forrester Wave evaluations,
customer experience, industry trends, new and emerging markets, and architecture. More

Imad Birouty
Director of Product Marketing
Teradata
Imad Birouty holds the position of Director of Teradata Product Marketing and is responsible for Teradata
software and hardware products including the Teradata Database, Teradata Platform Family, Teradata
QueryGrid, Teradata Unity, Tools and Utilities, and our In-Database Analytics. Prior to this, Imad led the Product
Management team responsible for the Teradata Platforms; setting product strategy and direction.

2015 Teradata

@nyuhanna
#inmemory

In-Memory Processing for High Performance


Analytics
Noel Yuhanna,
Principal Analyst, Forrester Research

August 11th 2015


4

2009
2014 Forrester
Forrester Research,
Research, Inc.
Inc. Reproduction Prohibited

Teradata Webinar

Business growth and speed are


changing the need for accessing data
faster than ever its a competitive
advantage.
5

2014 Forrester Research, Inc.

Data is the new .

Currency
Oil

Bacon
2015 Forrester Research, Inc.
6
2014 Forrester
Research,
2015
Research,
Inc.Inc.Prohibited
2014 Forrester
Forrester Research,
Inc. Reproduction

It has become critical for


every business!
6

However data explosion means more cost, slower


data access and increasing data challenges

social

Video

sensors

of data is on the
public net

Cloud

BIG DATA

IOT
7

2014 Forrester Research, Inc.

mobile

Challenge
Businesses think of analytics as a set of boring
historical reports and dashboards they dont
want yesterdays data tomorrow!
8

2014 Forrester Research, Inc.

We are moving from IT to Business Technology (BT)..


Self-service
Digital
business

Real-time

Lots of
data

BT
Batch
Limited
data
MIS

IT

Trend
Automation

9
2014
Forrester Research,
2015 Forrester
Research,
Inc. Inc.

Real-time data access requirements have grown.


Mobile devices we need data now!
Competitive pressure to act more quickly

Run OLTP and DW/Analytical apps faster


Need to act quickly with IOT/sensor data
New insights, advanced analytics that need data quicker

2015 Forrester Research, Inc.


10 Forrester
2014 Forrester
Research,
2015
Research,
Inc.Inc.

What technology is helping with real-time?


Falling memory (DRAM) prices from
over $100K/GB in 1990 to less than
$1/GB today.

Organizations are already running a few


Terabytes of in-memory and we are
heading petabytes by 2018+
2015 Forrester Research, Inc.
11 Forrester
2014 Forrester
Research,
2015
Research,
Inc.Inc.

400
300
200
100
0
2002
2002
2003
2004
2006
2006
2007
2008
2009
2009
2010
2011
2012

Support for SSD/Flash along with Disks

memory

Falling memory prices


8

Two tracks support various real-time analytical


needs
Faster access
Real-time

Batch

9
2015 Forrester Research, Inc.
12 2014,
2014Forrester
Forrester Research,
Inc. Inc.
Research,

Streaming Analytics
Software that can filter, aggregate, enrich, and analyze a
high throughput of data from disparate live data sources to
identify patterns to visualize business in real time, detect
urgent situations, and automate immediate actions

2015 Forrester Research, Inc.

13

2014 Forrester Research, Inc.

11

Distributed in-memory technology helps deliver


real-time high performance analytics
Business Insights

Operations

Analytics
DRAM

Clickstream

Distributed in-memory
Sensors

SSD/Flash

(Horizontal scale)

Logs

streaming

Disk

Tickers

Traditional sources
2015 Forrester Research, Inc.
14 Forrester
2014Forrester
Forrester
Research,
2014,
Research,
2015
Research,
Inc.Inc. Inc.

Apps
.

Database

Other sources
14

Use cases typically seen with in-memory processing


Age of the Customer personalization
IOT machine analysis, proactive maintenance
Real-time analytics various business insights

Operational intelligence
Fraud Detection risk management, online trading
Mobile Apps Hotel reservations, inventory tracking
And others
2015 Forrester Research, Inc.
15 2014,
2014Forrester
Forrester Research,
Inc. Inc.
Research,

15

Age of the customer is driving the need for a realtime data platform.

2015 Forrester Research, Inc.

16

2014 Forrester Research, Inc.

2014 Forrester Research, Inc. Reproduction Prohibited

16

17

Consumer personalization has become critical


for organizations to succeed

2015 Forrester Research, Inc.

17

2014 Forrester Research, Inc.

2014 Forrester Research, Inc. Reproduction Prohibited

17

18

Age of the Customer

How can you prevent


members from churning?
18

2014 Forrester Research, Inc.

IOT drives new types of use cases but requires a


real-time data platform
IOT

Engines

Machines
19

2014 Forrester Research, Inc.

2014 Forrester Research, Inc. Reproduction Prohibited

2015 Forrester Research, Inc.

Factories

Cisco predicts 50 billion


devices will be connected
by 2020.
Forrester estimates 30%
usage of IOT in
manufacturing, will double
by 2019.
19

IOT analytical Apps continue to grow rapidly


Fleet management Monitoring condition, location, and usage of vehicle fleets
Inventory management - Tracking inventory levels and managing operations
Facility management
Customer order and delivery tracking - Enabling customer visibility
Energy management - Monitoring, usage of water, electricity, and other resources
Smart products

Supply chain management - Managing supply chain relationships


Smart home management
Industrial asset management
Security and public safety monitoring or surveillance

2014
Forrester Research,
201520
Forrester
Research,
Inc. Inc.

20

What does in-memory mean to the business and IT?

Business

IT

21

Use in-memory platform to innovate and become a


disruptor, there are endless possibilities
Intensify customer digital experiences
Insist on real-time analytics for various use cases to
gain competitive advantage
Look closely at vendor solutions that can scale
Focus on tiered memory Dram, SSD/flash, disk
Expand in-memory to support more data and new realtime analytical use cases

2014 Forrester Research, Inc.

2014 Forrester Research, Inc. Reproduction Prohibited

21

Thank you
Noel Yuhanna
www.forrester.com

Twitter: @nyuhanna

2009 Forrester Research, Inc. Reproduction Prohibited

Teradata Intelligent Memory


Imad Birouty
Director, Teradata Product Marketing

Finding The Right Balance


Memory is 3,000x faster
than disk

Memory still 80x more expensive


than disk

Cost of memory is
decreasing

Not all data worth 80x premium

Memory per node is


increasing

96GB -> 256GB ->512GB


-> 768GB -> 1TB

24

Data Warehouse has wider


variety of data than OLTP systems

Not all data fits into memory


Blindly adding memory has
diminishing returns

Teradatas Approach To In-Memory


Advanced engineering to use memory
intelligently
Economical use of memory

Integrated into the Teradata Database


No separate DBMS to manage

Automated in-memory data management


Scale to the largest system sizes (2,048 nodes)
Database internal code changes to optimize
for in-memory processing
Improve overall system efficiency
25

In-Memory For
Your

Most Valuable
Data

26

Understanding Data Access Patterns


Performance of in-memory databases with disk economics
Data Temperature Profile Typical DW

Teradatas Approach
Hottest data in
memory/not all the
data
Integrated into
Teradata system

No need for separate


appliance

27

Teradata Intelligent Memory

28

Sophisticated
algorithms to track
usage, measure
temperature, and
rank data

Memory
Cache

Intelligent
Memory

Temporarily store data


required for current queries,
purges least recently used

Hottest data placed and


maintained in memory,
aged out as it cools

most
recently
used
data

most
frequently
used data
very hot in

cool out

Compliments FSG
cache
Dynamically
adjusts to new
query patterns

Advanced
Engineering For

Maximum
Performance

29

Smart Engineering To Get Most From Systems


Maximize System Throughput By Reducing Bottlenecks

Fast

Faster

Fastest

Goal
Goal
of Advanced
of In-Memory
In-Memory Systems Engineering
30

Engineering For Performance and System Efficiency


It Is More Than The Amount of Memory
Its about taking full advantage of modern CPU technology
Its about Improving memory bandwidth
Its about Improving processor cache effectiveness

Increasing memory improves Disk I/O


Improving the bandwidth and cache effectiveness
improves cost per instruction, throughput, and
response time
31

Pipelining & Advanced Use of Memory


Query Pipelining & New In-Memory Table Structures

With Pipelining

Without Pipelining

Improves Performance with fewer disk I/Os


Optimizes Memory Bandwidth
Improves CPU Throughput

Node

Node

Disk

Disk

32

New in-memory
table structures
hold data as
column partitioned
to reduce size and
store data in the
way the CPU
accesses it

Bulk Qualification & Vectorization

Memory Bandwidth and Memory Cache Effectiveness


Vector
CPUs
Instr

CPU

Vector List
Pointer

Instr

Pointer

Instr

Pointer

Instr

Pointer

Instr

Pointer

Instr

Pointer

Instr

Pointer

Instr

Pointer

Instr

Pointer

SIMD
Album_Table

SELECT * From Album_Table


WHERE Producer = 'Smith';

33 2015 Teradata

Column Partitioned Table

In-Memory Hash Join


Qualified Rows
held in InMemory Spool

Store Table

34

Stores that offer pharmacy services

Store Services

(based on Hash
Value bulk
qualification)

Row at a Time
Evaluation

CA

store_id

Vector-based
Evaluation

Los Angeles

CA

710

Orange

712

store_id

city

state

601

San Diego

CA

602

San Diego

701

store_id

Pharmacy

Tires

601

store_id

602

601

601

701

CA

602

602

710

Tucson

AZ

701

701

712

725

Scottsdale

AZ

710

710

725

726

Scottsdale

AZ

712

712

726

729

Phoenix

AZ

725

725

729

726

726

729

729

2015 Teradata

Teradata Intelligent Memory

Performance of
In-Memory
Economics of Disk
Storage
35

Questions?
Ask via the Chat function now

36

2015 Teradata

37
37

Você também pode gostar