SUGMENA In-Memory

wispxylopolistInternet και Εφαρμογές Web

7 Αυγ 2012 (πριν από 5 χρόνια και 8 μέρες)

469 εμφανίσεις

In
-
Memory Analysis with HANA and Sybase
IQ
High
-
Speed Data
Warehousing

Waldemar

Adams

Head of BI Center Of Excellence

SAP EMEA


May 2011

2

©
2011 SAP AG. All rights reserved.

In
-
Memory Computing

Technology
that allows the processing of

massive quantities of real time data

in the main memory of the server

to provide
immediate results
from

analyses and
transactions

3

©
2011 SAP AG. All rights reserved.

“Business Analytics” is Nothing New

The “What” doesn’t fundamentally change


but the “How” does

4

©
2011 SAP AG. All rights reserved.

photo by
Jurvetson

(
flickr
)

The End of an Era

5

©
2011 SAP AG. All rights reserved.

Progress

Transistors: much simpler, much smaller, much
cheaper, more reliable, no warm up, much
faster.

Integrated circuits: miniaturization added to all
the existing benefits, enabled unthought
-
of
possibilities

Vacuum tubes: slow,
expensive, fragile

6

©
2011 SAP AG. All rights reserved.

Reporting

“Typical” Business Intelligence Today

Slow

Painful

Expensive

Operational Data Store

Data Warehouse

Indexes

Aggregates

Data

Business Applications

Copy

ETL

Calculation Engine

Business Intelligence

Query Results

Query

Slow

Painful

Expensive

Operational Data Store

Data Warehouse

Indexes

Aggregates

Data

Business Applications

Copy

ETL

Calculation Engine

Business Intelligence

Query Results

Query

Data

Marts

7

©
2011 SAP AG. All rights reserved.

The Analyst View

8

©
2011 SAP AG. All rights reserved.

In
-
Memory = Speed Of Thoughts...

0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
Cache/on CPU
RAM/Memory
Flash Drive
Access time in CPU cycles
Access time in
CPU cycles
In
-
Memory vs. SSD (
Solidstate
/Flash
-
Drives)

Min
Max
0
1000000
2000000
3000000
4000000
5000000
6000000
7000000
8000000
9000000
10000000
Cache/on CPU
RAM/Memory
Harddrive (seek)
Access time in CPU cycles
In
-
Memory vs. Harddisk

9

©
2011 SAP AG. All rights reserved.

In
-
Memory Computing

Disk is 1Mx slower than direct memory: like a chef doing his shopping on mars

10

©
2011 SAP AG. All rights reserved.

In
-
Memory Computing Costs dramatically decreased

Burj

Al Arab, Dubai, 210
m rooftop

Cost of 1 Mb of
memory in 2000:


1

And in 2010: just
<

1 cent…

11

©
2011 SAP AG. All rights reserved.

Row
-
Based Data

Wasted space,
and a full scan to
aggregate any
particular field

12

©
2011 SAP AG. All rights reserved.

Column Data

More efficient data storage, better compression, faster queries

13

©
2011 SAP AG. All rights reserved.

Combine Them: In
-
Memory, Column Database

Now you can store the entire supermarket right by your cooking counter


14

©
2011 SAP AG. All rights reserved.

In
-
Memory Computing

Operational Data Store

Data Warehouse

Indexes

Aggregates

Data

Business Applications

Copy

ETL

Calculation Engine

Business Intelligence

Query Results

Query

Up to 1,000x faster

No optimizations required

More data in less space

Faster BI

Data

Marts

15

©
2011 SAP AG. All rights reserved.

Data Warehouse

Data Warehouse

Column Databases

Operational Data Store

Data Warehouse

Data

Business Applications

Copy

ETL

Calculation Engine

Business Intelligence

Query Results

Query

Up to 1,000x faster

More efficient data storage

Better compression

More data in less space

Faster BI

16

©
2011 SAP AG. All rights reserved.

Data Warehouse

Analytic Appliance

Operational Data Store

Data

Business Applications

Copy

ETL

Business Intelligence

Query Results

Query

Up to 1,000x faster

Massively parallel, optimized for hardware

Move calculation engine to the database: in
-
memory predictive analytics

Include a BI server

Less data

movement, f
aster BI

Analytic Appliance

Calculation Engine

17

©
2011 SAP AG. All rights reserved.

Real
-
Time Data

Operational Data Store

Copy

ETL

Less latency

Row
-
level data in DW

Incremental updates, update only

(audit in time)

Real
-
time data integration, data quality

So why have a separate operational data store?

Data

Business Applications

Analytic Appliance

Business Intelligence

18

©
2011 SAP AG. All rights reserved.

The New Generation of Analytic Platforms

Copy

Business Applications

Analytic Appliance

Business Intelligence

In
-
memory, column

data store, on massively parallel appliance

Integrate personal, structured and unstructured data

Put into the cloud, distribute across mobile devices

Include collaboration


Data

19

©
2011 SAP AG. All rights reserved.

Virtuous Circle of Technology

In
-
Memory

Columnar
Databases

Hardware
Acceleration

Calculation
Engine

Columnar storage
increases the
amount of data that
can be stored in
limited memory
(compared to disk)

Column databases
enable easier
parallelization of
queries

In
-
memory processing
gives more time for
relatively slow updates
to column data

In
-
memory allows
sophisticated calculations
in real
-
time

Hardware acceleration
makes sophisticated
calculations like
allocations possible

Each technology works well
on its own, but combining
them all is the real
opportunity


provides all of
the upside benefits while
mitigating the downsides

Delivering on this Vision

23

©
2011 SAP AG. All rights reserved.

HANA?
How

it

started

...

24

©
2011 SAP AG. All rights reserved.

SAP
H
igh
-
Performance
AN
alytic

A
ppliance (HANA)

High
-
Performance


Innovative
data replication
& ETL data services


Native connect of SAP BO BI

front
-
ends plus

multiple interfaces for real
-
time analysis

Analytic


Tools
for

data modeling
, data &

life
cycle
management


Scalable and rich BI
platform & BI clients with

SAP
BusinessObjects

BI

Appliance


In
-
Memory
software

bundled with hardware
delivered from our hardware partner

(
HP, IBM, Fujitsu
)

In
-
Memory Computing
Engine

Admin and Data Modeling

Real

Time Replication
Services

Data Integration Services

In
-
Memory

Computing

Calculation
and
Planning
Engine

SAP
NetWeaver


BW

SAP

Business
Suite

3
rd

Party

Other
Applications

SAP
BusinessObjects

MDX

SQL

BICS

Data Management Service

25

©
2011 SAP AG. All rights reserved.

154,000

B2B customers

SAP HANA in Action at a CPG Company

1.8M

1,000

70,000

rows of data

t
arget customers

collection
notices generated

13

seconds

77
minutes

Standard System

In
-
Memory
System

(with 4 CPU Xeon)

356x

faster

26

©
2011 SAP AG. All rights reserved.

SAP Strategy for In
-
Memory


EXPAND PARTNER ECOSYSTEM


Partner
-
built applications, Hardware partners



CUSTOMER CO
-
INNOVATION


Design with customers


TECHNOLOGY INNOVATION


BUSINESS VALUE


Real
-
Time Analytics, Process Innovation, Lower TCO


GUIDING PRINCIPLES


INNOVATION WITHOUT DISRUPTION


New Capabilities For Current Landscape


HEART OF FUTURE APPLICATIONS


Packaged Business Solutions for Industry and Line of Business

27

©
2011 SAP AG. All rights reserved.

HANA, BWA, Sybase IQ

and how they complement

BWA

Speed up BW

corporate EDW


SAP HANA 1.0 is based on In
-
Memory technology to host the data (RAM).

Data structure is both column
-
based but also row
-
based.


SAP NetWeaver Business Warehouse Accelerator (BWA) is based on In
-
Memory technology

to host the data (RAM). Data structure is column
-
based.

Moving forward, BWA evolves into SAP HANA as SAP
i
n
-
memory computing becomes the
primary persistence mechanism for BW (with HANA
SP3)


Sybase IQ uses column
-
based structure to organize data representation and conventional storage
technology to host the data (
Harddrives
, RAID, SAN)

Sybase IQ

Analytical Data
Marts for non
-
SAP sources

HANA 1.0

SAP ERP

Operational
Reporting

28

©
2011 SAP AG. All rights reserved.

2011 Guidelines for In
-
Memory

HANA 1.0 to All SAP Customers

Real
-
Time Operational Reporting for SAP

Agile Data Mart for Any
Data

Sybase IQ to non
-
SAP data Customers

Agile, Analytics Data Marts for Any Data

All 2011, leverages
all SAP
BusinessObjects

BI and EIM
(
DataQuality
!) suite

Explorer/BWA to Existing SAP BW Customers

SAP BW Performance Boost

All 2011

HANA 1.0

30

©
2011 SAP AG. All rights reserved.

Information is the foundation of any business
transaction or decision

33

©
2011 SAP AG. All rights reserved.

SAP
BusinessObjects

Data
Services for HANA

SBO BI Clients

Hana

Appliance

Calc & Planning

Engine

Data Layer

NewDB

Data Services

Hana

Studio

Business

Suite

SAP BW

3
Rd

Party

Optional Components

High Performance


Highly
-
scalable engine to move large
volumes of data into HANA


Integrated with HANA’s bulk
-
load
interfaces

Wide Connectivity


Data and Metadata connectivity to all
major enterprise data sources


Native, fast connectivity to
Applications, RDBMS, Files

Powerful Transformations


Built
-
in transformations for data
quality



Support for non
-
relational data
formats including text and XML







Leading data integration solution for high
-
performance batch loading of data into
HANA

Data Services can be used for scheduled (non
-
real
-
time)
loading of data from different sources into HANA.
Replication Server is used for real
-
time movement of data
into HANA

34

©
2011 SAP AG. All rights reserved.


Oracle


DB2


Sybase & IQ


SQL Server


Informix


Teradata


ODBC


MySQL


Netezza


JD Edwards


Oracle Apps


PeopleSoft


Siebel


Salesforce.com


SAP
NetWeaver


BW


SAP R/3


ABAP


BAPI


Idoc


Extractors*


Text delimited


Text fixed width


EBCDIC


XML


Cobol


Excel


HTTP


JMS


SOAP

(Web Services)


ADABAS


ISAM


VSAM


Enscribe


IMS/DB


RMS


Both direct and
change data

Data Services: Enterprise
-
Wide Data Acces
s

Broad connectivity to databases, applications, legacy, file formats and unstructured data


Any text file type


32 languages

Databases

Applications

Files/Transport

Mainframe

(with partner)

Unstructured Data

*
Delta

extractor support available in certain versions of
NetWeaver

only

35

©
2011 SAP AG. All rights reserved.

Real
-
time Analytics


Real
-
Time Operational
Analytics


Agile LOB Data Marts


Accelerated Analytics with
BI 4.0



One Store for Data and Analytics


HANA only persistence layer for
SAP Business Suite


Business Suite runs on HANA
future release


SAP Business Suite optimized
for In
-
Memory


Eliminate
3rd
party

database


Flexible real time analysis
of operations at non
-
aggregated level


Real
-
Time operational
planning, simulation and
forecasting: link to execution


Reduced landscape
complexity


Value chain transformation

Capabilities

Benefits

Next generation applications


SAP BW fully running on HANA
SP3


HANA SP3 platform for IM Apps


SBOP 4.x (Aurora) unified
modeling with Hana


Industry and LOB Business
Analytics Solutions “BAS”

Q4
2010

“Renovation”

HANA 1.0

“Innovation ”

HANA
SP3

“Transformation”

HANA
future release

SAP In
-
Memory Strategy

Product Strategy

36

©
2011 SAP AG. All rights reserved.

Analyst View: Gartner

37

©
2011 SAP AG. All rights reserved.

SAP HANA: Value Proposition

Addressing Key Business Drivers


1.
Real
-
Time Decision Making


Fast and easy creation of ad
-
hoc views on business


Access to real time analysis

2.
Accelerate Business Performance


Increase speed of transactional information flow in areas such as planning,
forecasting, pricing, offers…

3.
Unlock New Insights


Remove constraints for analyzing large data volumes
-

trends, data mining,
predictive analytics etc.


Structured and unstructured data

4.
Improve Business Productivity


Business designed and owned analytical models


Business self
-
service


reduce reliance on IT


Use data from anywhere

5.
Improve IT efficiency


Manage growing data volume and complexity efficiently


Lower landscape costs

Speed

Scale

Flexible

Thank You!


w
aldemar.adams@sap.com