50
© 2012 SAP AG. All rights reserved. 1 SAP In-Memory Computing SAP HANA Deep Dive Chris Bullock 10/9/2012

HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

Embed Size (px)

Citation preview

Page 1: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 1

SAP In-Memory Computing

SAP HANA Deep Dive

Chris Bullock10/9/2012

Page 2: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 2

Disclaimer

This presentation outlines our general product direction and should not be

relied on in making a purchase decision. This presentation is not subject to

your license agreement or any other agreement with SAP.

SAP has no obligation to pursue any course of business outlined in this

presentation or to develop or release any functionality mentioned in this

presentation. This presentation and SAP's strategy and possible future

developments are subject to change and may be changed by SAP at any

time for any reason without notice.

This document is provided without a warranty of any kind, either express or

implied, including but not limited to, the implied warranties of

merchantability, fitness for a particular purpose, or non-infringement. SAP

assumes no responsibility for errors or omissions in this document, except

if such damages were caused by SAP intentionally or grossly negligent.

Page 3: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 3

Big John

Page 4: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 4

Willis Tower – 1,451 ft Big John – 1,127 ft

Page 5: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 5

Burj Khalifa – 2,723 ft

Page 6: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 6

� An in-memory data management platform

� “Real Real-time” performance for huge data volumes

� Current focus: business analytics

� Technology foundation for new in-memory enterprise applications

� Delivered as a vertically integrated appliance via SAP HW partners

� Fastest growing productin SAP’s 40-year history

What is SAP HANA?

Speed

Scale

FlexibilityFlexibility

SpeedSpeed

Scale

Flexibility

Page 7: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 7

Cloud

SAP’s Business Strategy

Business Suite (ERP)

Business Analytics (BO, BW)

mobile

Data platform (SAP HANA)

HANA is at the heart of

SAP’s vision and strategy

Credit for illustration idea: Maarten de Vries, SAP

# 1, worldwide market share, 2010

# 1, worldwide market share, 2011

Gartner magic quarter leader

Fastest growing SAP product ever

15m paying cloud

subscribers

Page 8: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 8

SAP HANA: Business Value Proposition

Make Decisions in Real-time

Access to real time analysis; fast and easy creation of ad-hoc business statistics

Accelerate Business Processes

Increase speed of information processes such as planning, forecasting, pricing, offers…

Unlock New Insights

Remove constraints for analyzing massive data volumes, trends, data mining, predictive analytics…

Structured and unstructured data

Improve IT Efficiency

Manage growing data volume and complexity withlower cost of ownership

Speed

Scale

Flexibility

Page 9: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 9

A Combined and Simplified Architecture

In-Memory

Row +

Column Database

Massively Parallel

Processing

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

Row database fast transactional processing

In-memory processing gives more time for

relatively slow updates to column data

In-memory allows sophisticated calculations

in real-time

MPP optimized software enables linear performance scaling

making 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

Page 10: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 10

Sportsmart Demo

Image: Renjith Krishnan / FreeDigitalPhotos.net

Page 11: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 11

Technology Trends

Image: Renjith Krishnan / FreeDigitalPhotos.net

Page 12: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 12

Moore’s Law: DRAM Pricing

Source: OBJECTIVE ANALYSIS http://blogs-images.forbes.com/jimhandy/files/2011/12/DRAM-GB-Price.jpg

1 TB ~ $10K

100x

1000x

10x

Pri

ce

pe

r G

Byte

Multi-Terabyte servers are now completely affordable!

Page 13: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 13

Moore‘s Law: CPUs

2002

1 core32 bits4MB

2007

2 cores2 CPUs per serverExternal Controllers

8 cores -16 threads / CPU4 CPUs per serverOn-chip memory controlQuick interconnectVM and vector support64 bits; 256 GB - 1 TB

2010

More cores, bigger caches16 ... 64 CPUs per server Greater on-chip integration(PCIe, network, ...)Data-direct I/OTens - hundreds of TBs

2013

Images: Intel, Danilo Rizzuti / FreeDigitalPhotos.net

Page 14: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 14

Database Evolution

� In the eighties, we had simple, general purpose Database Servers…

Client Application

DatabaseServer

t2000 201019901980

Page 15: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 15

Database Evolution

� Over time, a separate “application server” tier, processing the data from the DB, was introduced; the web popularized this architecture

Database Server

App Server

Client(web)

t2000 201019901980

Page 16: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 16

Data

transformation and aggregation

Database Evolution

� Bigger and bigger datasets forced the introduction of separate, specialized “Data Warehouses”, used for analytic processing (OLAP)

Transaction Server

App Server

Client(web)

AnalyticClient

App Server

Data Warehouse

t2000 201019901980

Page 17: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 17

Database Evolution

� Using in-memory computing, SAP HANA offers a vision of simplicity and integration, with most of the data-intensive operations pushed into the data layer

DataClient

App Server

SAP HANA

t2000 201019901980

Page 18: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 18

Elements of HANA Technology

Image: Renjith Krishnan / FreeDigitalPhotos.net

Page 19: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 19

The New Challenges of In-memory Computing

� Challenge 1: Parallelism! Take advantage of tens, hundreds of cores� Challenge 1: Parallelism! Take advantage of tens, hundreds of cores

� Challenge 2: Data locality!

Yes, DRAM is 100,000 times faster than disk…

But DRAM access is still 4-60 times slower than on-chip caches

� Challenge 2: Data locality!

Yes, DRAM is 100,000 times faster than disk…

But DRAM access is still 4-60 times slower than on-chip caches

Page 20: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 20

SAP HANA Design Goals

Hardware innovation leads to software innovation

• In-memory: no disk access during read (updates are logged/persisted to disk)

• Highly parallel execution

• Cache-aware memory organization

Multi-engine data platform: beyond SQL

� Relational (Row and Column), Text, Graph, …

� Application-specific object models

� Embedded development environment (scripting)

� Business Function Library, intrinsic plan operators

� SAP Application server integration

Simplified System Architecture

• Reduced DB administration

• Integrated ERP transactions (roadmap)

Image: Renjith Krishnan / FreeDigitalPhotos.net

Page 21: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 21

Additional SAP HANA Design Goals

Support for up to very large data-sets

• Data partitioning and Data Distribution

• Up to ten servers (20TB physical, ~500B records) tested

Failure recovery and High Availability

� Persistence, Redo-log, save-points

� Backup / Restore (go back in history)

� Hot standby, lazy table-load

� Disaster recovery (clusters)

Other

• Planning Engine, Predictive Analysis

• Application-specific object repository

• Multi-tenancy and cloud deployment (future)

Image: Renjith Krishnan / FreeDigitalPhotos.net

Page 22: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

Insert only on change

Column and row store

++

No aggregatesMinimal

projections

Partitioning

Analytics onhistorical data

Single andmulti-tenancy

SQL interface on columns & rows

Reduction oftiers / layers

x

In-memoryCompression

Multi-core/parallelization

DynamicExtensibility

++ ++ ++

Active/passive& data aging

PAA

Bulk load

++

++

++++

T

Text Retrieval & Exploration

Multi-threading within nodes

Map reduce No diskGroup Key

t

SAP HANA Building Blocks

SQL

In-memory Apps

Page 23: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 23

� DBs typically use a row-based storage; SAP HANA supports rows, but is optimized for column-order data organization

Order Country Product Sales

456 France corn 1000

457 Italy wheat 900

458 Italy corn 600

459 Spain rice 800

Column and Row Store

456 France corn 1000

457 Italy wheat 900

458 Italy corn 600

459 Spain rice 800

456

457

458

459

France

Italy

Italy

Spain

corn

wheat

corn

rice

1000

900

600

800

Row order organization

Column order organization

Single-record access:SELECT * FROM SalesOrdersWHERE Order = ‘457’

SQL

Single-scan aggregation:SELECT Country, SUM(sales) FROM SalesOrders WHERE Product=‘corn’ GROUP BY Country

ΣΣΣΣ

Page 24: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 24

Order Country Product Sales

456 France corn 1000

457 Italy wheat 900

458 Spain rice 600

459 Italy rice 800

460 Denmark corn 500

461 Denmark rice 600

462 Belgium rice 600

463 Italy rice 1100

… … … …

Columnar Dictionary Compression

� Dictionary per column

� Uses data-driven fixed-length bit encodings

� Operations directly on compressed data, using integers

� More in cache, less main memory access

1 Belgium

2 Denmark

3 France

4 Italy

5 Spain

1 3

2 4

3 5

4 4

5 2

6 2

7 1

8 4

… …

1 7

2 5,6

3 1

4 2,4,8

5 3

Logical Table

Dictionary5 entries, so need 3 bits to encode!

Compressed column

(bit fields)Inverted

indexDictionary

Where was order 460?

Which orders in Italy?

Page 25: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 25

Columnar Run-length Encoding

� Compress repeated values in column memory

� Works best on sparse, sorted columns

� Other encodings in other cases

Order Country Product Sales

456 France corn 1000

457 Italy wheat 900

458 Spain rice 600

459 Italy rice 800

460 Denmark corn 500

461 Denmark rice 600

462 Belgium rice 600

463 Italy rice 1100

… … … …

1 Belgium

2 Denmark

3 France

4 Italy

5 Spain

3

4

5

4

2x2

1

4

Logical Table Country

1 corn

2 wheat

3 rice

1

2

2x3

1

3x3

Product

Page 26: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 26

Dramatic Simplification

Page 27: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 27

How HANA is Transforming Customers

Image: Renjith Krishnan / FreeDigitalPhotos.net

Page 28: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 28

This image cannot currently be displayed.

This image cannot currently be displayed.

This image cannot currently be displayed.

This image cannot currently be displayed.

This image cannot currently be displayed.

This image cannot currently be displayed.

This image cannot currently be displayed.

This image cannot currently be displayed.

This image cannot currently be displayed. This image cannot currently be displayed.

This image cannot currently be displayed.

This image cannot currently be displayed.

Over 400 New HANA Customers …

Page 29: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 29

HANA – Usage Scenarios Today

Operational Data Mart / Application Accelerator

• Flexible Real-Time Analytics/Reporting

• Accelerated SAP Applications

• Rapid Deployment Solutions for Quick Deployment

Agile Data Mart

• Enhance Existing Data Mart and Data Warehouse Investments

• Data Acquisition and Integration from Any Source

• Real-Time Consolidated Reporting/Analytics

SAP BW on HANA

• Dramatically Improved Performance

• Simplified Administration & Streamlined Landscape

• Unlock Data Across the Enterprise

• Preserve BW Investment without Disruption

Page 30: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 30

Southern California Edison Business Drivers

OperationalImprovement

Next GenerationAnalytics

• Faster reporting• Faster data loading• Lower TCO• Reduced maintenance costs• Reduced development costs

• Faster analytics• Modeling flexibility• Near real-time data

replication• Calculation engine and

built-in functions• Big Data footprint• New applications—

potential examples:• Smart meter analytics• Power outage management• Power procurement• Predictive analytics

Standalone HANA DW HANA

Page 31: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 31

The Southern California Edison Business Intelligence Journey

DW

2007 (1)DW

2009 (2)DW

2011 (3)DW HANA

2012 (4)Standalone HANA

2012 (5)Integrated HANA

2014+

Legacy DBWithout

BI

Legacy DBWithBI

Legacy DBWithBI

HANAWithBI

Calc Engines Enterprise Apps, DW,and Standalone

Integration

Baseline Faster Report Speed

• BI 2.5x fasterreporting

Faster Report Speed

• BI verystable

• Reports are faster:Avg (20%)

Legacy DB repl withHANA

• Reports are faster5.0 times

• Delta Data Loads aremuch faster (3.2x)

HANA Standalone Database

• Optimized calculation engines with parallel processing

• Radical Improvement

HANA Standalone Database

• HANA environmentsvirtually or physically merge

• Reports:• Avg 90 sec• CRM 400 sec

• Data Loads 15 hrs

• Reports:• Avg 40 sec• CRM 161 sec

• Data Loads 15 hrs

• Reports:• Avg 32 sec• CRM 23 sec

• Data Loads 15 hrs

• Reports:• Avg 6.4 sec • CRM 4.6 sec

• Data Loads 4.6 hrs

• Analytics: • Telecom CRM 55:1• Smart Meter

Analytics 40:1

• More agile• Seamless• Less costly

development

Performance Improvements

0

10

20

30

40

50

60

70

80

90

100

1 2 3 4 5

Reports

Data

Analytics (New Capabilities)Operational Improvement (DW HANA)

Relative estimate

Projected

Based on Pilot Results

Page 32: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 32

HANA – Pilot

Validate:

• Compression• Data loading speed• Report response time• Back-up and restore• Security

W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 W13 W14 W15 W16 W17

Task Name 1/9/2012 1/16/2012 1/23/2012 1/30/2012 2/6/2012 2/13/2012 2/20/2012 2/27/2012 3/5/2012 3/12/2012 3/19/2012 3/26/2012 4/2/2012 4/9/2012 4/16/2012 4/23/2012 4/30/2012

HANA Pilot Program

Project Management

Pilot Planning

Production Planning

System Setup

Hardware

Software

Data Migration

DW HANA

HANA Standalone

DW HANA Enhancement

DW HANA Optimization

Data Modeling

Design

Build

Migrate and Validate

Security Setup

DW HANA

HANA Standalone

Report Development & Test

DW HANA

HANA Standalone

Explorer Testing

SAS Validation

SLT Validation

Backup Restore

Page 33: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 33

Pilot ResultsComparing DW HANA vs DW on Legacy Database

0

500

1000

Legacy HANA

Data Loading - Full

DSO to Cube

DSO to DSO

DSO Activation

PSA to DSO

Source to PSA

5.2 x

0

20

40

60

Legacy HANA

Data Loading - Delta

DSO to Cube

DSO to DSO

DSO Activation

PSA to DSO

Source to PSA

3.2 x

5 x 2 x

7.5 x

0

20

40

Response Time (Simple)

Response Time

0

10

20

30

40

Response Time (Complex)

Response Time

0 50 100 150 200 250

Legacy DB

HANA

Data Compression (column store)

Cube DSO Master Data

Projected for Production = 5.7 x

Page 34: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 34

HANA – Production

Goals:

• Migrate all of DW Legacy DB into DW HANA• Incorporate BCS and IP into DW HANA• Reduce nightly batch loading• Improve reporting performance• Migrate enterprise BI • Create the ability to handle “Big Data” analytics• Leverage built-in calculation engines• Create one or more new applications in Standalone HANA

W1 W3 W5 W7 W9 W11 W13 W15 W17 W19 W21 W23 W25 W27 W29 W31 W33

Phase 5/1/2012 5/15/2012 5/29/2012 6/12/2012 6/26/2012 7/10/2012 7/24/2012 8/7/2012 8/21/2012 9/4/2012 9/18/2012 10/2/2012 10/16/2012 10/30/2012 11/13/2012 11/27/2012 12/11/2012

Planning

Project Management

Pre-Trial

Development/ Unit Test

Reg Testing Cycle 1

Reg Testing Cycle 2

User Acceptance

Performance Testing

Production Cutover

Post Production

Stabilization

Page 35: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 35

Current Environment – DW Legacy DB

Sources

ECC

CRM

SRM

Others

Ext

racto

rs

EDW

DSO

DSO

DSO

INT 1

DSO

DSO

DSO

INT 2

DSO

DSO

DSO

Data Mart

PC PC PC PC CubeCube

CubeCube

CubeCube

Multi P

rovid

ers

BeX Universe

Unv

Unv

Unv

BEx

BEx

BEx

MDX

Reports

Crystal Reports

WebI

Xcelsius

BICSBICS

BWA

Legacy DB

New 2012 Environment – DW HANA, BI 4.0, and Data Services 4.0

Sources

ECC

CRM

SRM

Others

Ext

racto

rs

EDW

DSO

DSO

DSO

INT 1

DSO

DSO

DSO

INT 2

DSO

DSO

DSO

PC PC PC

Multi P

rovid

ers

BeX Universe

Unx

Unx

Unx

BEx

BEx

BEx

MDX

Reports

Crystal Reports

WebI

Xcelsius

BICSBICS

HANA

Business Intelligence Development Process

Page 36: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 36

HANA Standalone Environment – Standalone HANA

Sources

ECC

CRM

CRM

Teradata

Ext

racto

rs

Universe

Unx

Unx

Unx

MDX

Reports

Crystal Reports

WebI

Xcelsius

HANA

OracleData Services

Ext

ract

Tra

nsfo

rm

EDW

T1

T2

T3

T4

Load

Data Mart

T1

T2

T3

T4

T5

HA

NA

Mo

de

ler

AttributeView

AnalyticalView

CalculationView

Business Intelligence Development Process

Page 37: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 37

Storage Sizing - Projection

Object TypeCurrent Size in Legacy DB

Tighten DW House Keeping

Clean-up System Tables

Decommission Unused Objects

Reduce Layers

Near lineOld Data

PSA 5,842 100 100 100 100 100

Legacy DB Overhead 6,141 - - - - -

Change Log 4,174 - - - - -

DSO 1,487 1,487 1,487 1,312 1,201 1,201

System Tables 1,167 1,167 300 300 300 300

Cube 708 708 708 648 591 141

Master Data 408 408 408 408 408 408

Temp 7 7 7 7 7 7

Total 19,934 3,877 3,010 2,775 2,607 2,157

6 x 3322 646 502 462 434 359

* All Numbers are in GB

HANA DB Compression

Page 38: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 38-

Standalone HANA – Potential Opportunities

Smart Meter Analytics• Improve customer energy efficiency• Improve campaign effectiveness• Understand consumption behavior

Power Procurement• Lower short-term power purchases• Better forecasting• Introduce new pricing options

Outage Management• Reduce response time and outage costs• Increase customer satisfaction• Quickly analyze up-to-date outage information

Predictive Analysis• Forecast outages, usage patterns, and cost• Model customer segmentation• Rapidly analyze advanced statistical models

Page 39: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 39

Colgate-Palmolive begins budget/planning roll out

�Colgate uses a financial

application for their entire

budget and planning process.

The process starts with laying

out the budgets and plans for

future periods and then actuals

come into the system

�They are limited in what

analysis they can perform (due

to data volumes, aggregation,

hierarchies, etc.)

�Need a cross-enterprise

solution. Colgate started with

their Hills Pet Nutrition business

�Hills Metrics: 450M records

from ERP system, 600GB data.

100k-200k change records/day

�Had to be non-disruptive:

� If Colgate had to be on the latest

software stack they could not

go-live before End 2012

� Zero training or UI impact to

users

Reporting by Customer/Product

now available for 1st time

Flexible customizing of financial

application is key, (e.g., switch on

and off for specific users, reports,

operating concerns) allowing a

smooth and non-disruptive go-live

with minimal risk

Challenge SolutionConditions

Page 40: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 40

Accelerate Business Performance

Large CPG company wants to analyze all their POS of data to predict demand

Target -stock shelves with 48 hour turn-around

Data Set is 460 Billion records (40 Terabytes)

Unable to analyze data using current database platform

10 HANA blades with 500GB per blade (5TB) & 2TB SSD Storage

SAP BusinessObjects Explorer

Significant data compression

20x Faster Analysis with 200x Better Price/Performance

Moved from 5 days down to 2 days for shelf turnaround

Eliminates out of stock scenarios during promotions

Challenge ResultsSolution

SAP HANA for Data Intensive Point of Sale Analysis

Page 41: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 41

Consumer Products CompanyFinancials with SAP HANA

15-60 minutes to generate summary

profitability report

2.9 seconds for600 million records

Drill-down to detail

Analyze any SKU, product family, region, time period …

Page 42: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 42

Global Professional Services Firm— the “Finder” Application

In the sales process, the current approach to

understanding the depth of the relationship

with a customer or prospect is a global email

(spam)

To transform this, need to collect data from

several SAP and non-SAP sources:� HR, FI, CRM

� Recruiting DB

� Alumni DB

� Linked In, etc.

Once collected, you need to search based on

existing/historical relationship with the

customer or prospect

Then score the results based on an algorithm

measuring the depth of the relationship (e.g.,

senior-level, multi-year)

To support sales pursuits, respond on mobile

devices to queries about relationships, then

send targeted messages to the key relationship

owners to help in pursuit

Page 43: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 43

DW on HANA, Major Beer Brewer

� DW Source was 4TB – HANA was 400GB (10x compression is

fairly standard, not allow for dynamic memory demands of

HANA)

� Data load was 10x faster (specifically, the portion that does

transformation and activation inside DW, not extraction from

source data which happens outside HANA)

� Queries on HANA 3X faster than previous in memory solution

� Upgrade and Unicode migration was manageable – about 4

technical days to complete

� Adding fields & attributes, is done in minutes (max), no re-

aggregation or indexing required

� Of 30 DW resources, they feel they will be able to reassign 6

people to other task. 6 redeployed staff is significant

Page 44: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 44

T-Mobile Initial Production Scenario

New Marketing program initiative—to drive an aggressive Marketing initiative for targeted campaigns and offers

� Offers are available for approximately 21 million customers

� Presented via retail stores, customer care centers and eventually via SMS. Each offer has a limited lifespan

� Goal is to increase adoption rate & profitability, customer retention

� Marketing Operations team needs to collect, analyze and report results of these campaigns/offers very quickly and with great flexibility

� Combine subscriber, marketing data, and POS data (20 tables from four sources). Initial load of nine months of data: inbound takes table ~82 million records; outbound offers table ~600 million records.

�Lengthy data load times

�Need access to details of customers actions, stop aggregating data

�Two primary data sources:

�(1) Enterprise data warehouse; and

�(2) Marketing campaign application

�Teradata sourced—so used Data Services ETL. Business Objects reporting tools in use for both the operational reports and ad hoc reporting

�Short fuse on project time frame

� “50x improvement in analytics in offer optimization…”

� “Marketing now gets their reports in 3 hours rather than 1 week and can recalibrate offers real-time”

� “We could not find anything that we already owned to solve this problem including Greenplum, Netezza, Teradata….they were too expensive or too risky.”

Challenge SolutionConditions

Page 45: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 45

Medidata & Big Data

Leading Provider of SaaS technology solutions that drive efficiencies across clinical research processes for Life Sciences companies

Data Volumes

� Over 25,000 protocols

� Over 250,000 site relationships

� Over 2,000,000 patients

� Billions of rows of transactional data for every clinical trial activity

Today

• Use HANA for specific analytics and benchmarks

• Dashboards generated in real-time

• Dashboards that can be “explored” in real-time with interactive visualization

• Tomorrow Leverage for broader cross-functional analytics

Actionable Analytics to help our customers optimize operations

First experience, Complex Query:

� Inner & Outer Joins across 17 tables

� Large Tables (millions of rows each)

47 minutes

1.78 Sec

Page 46: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 46

Experience SAP HANA Use Cases

Page 47: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 47

In Summary

� HANA is a disruptive in-memory data management platform, demonstrating a breakthrough in big-data analytic performance

– Customers already use SAP HANA to gain “real real-time” business insights into massive amounts of data

� SAP HANA represents a new paradigm for developing data-intensive applications

– HANA is vertically integrated with analytic applications (such as SAP BW) and delivered as an appliance with top HW partners

– SAP plans to release several new in-memory applications

– SAP intends to integrate its ERP business applications with HANA

Speed

Scale

Flexibility

HANA is at the heart of

SAP’s vision and strategy

Page 48: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

Thank You!

SAP HANA

Page 49: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 49

References

� SAP HANA on YouTube: http://www.youtube.com/watch?v=kwu5fndwz9Ahttp://www.youtube.com/watch?v=r3cw0yRKjsQ

� Excellent recent article:Faerber et al, SAP HANA Database - Data Management for Modern Business Applications; SigMOD Record December 2011, (Vol. 40, no. 4).http://www.sigmod.org/publications/sigmod-record/1112/pdfs/08.industry.farber.pdf

� More on HANA from SAP:www.sap.com/hana

� www.experiencesaphana.com

Page 50: HANA Deep Dive - BI / DW Insiderbi-insider.com/wp-content/uploads/2012/10/HANA_Deep_Dive.pdf · SAP HANA Deep Dive Chris Bullock 10/9/2012 ... business statistics ... Operational

© 2012 SAP AG. All rights reserved. 50

No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of SAP AG. The information contained herein may be changed without prior notice.

Some software products marketed by SAP AG and its distributors contain proprietary software components of other software vendors.

Microsoft, Windows, Excel, Outlook, PowerPoint, Silverlight, and Visual Studio are registered trademarks of Microsoft Corporation.

IBM, DB2, DB2 Universal Database, System i, System i5, System p, System p5, System x, System z, System z10, z10, z/VM, z/OS, OS/390, zEnterprise, PowerVM, Power Architecture, Power Systems, POWER7, POWER6+, POWER6, POWER, PowerHA, pureScale, PowerPC, BladeCenter, System Storage, Storwize, XIV, GPFS, HACMP, RETAIN, DB2 Connect, RACF, Redbooks, OS/2, AIX, Intelligent Miner, WebSphere, Tivoli, Informix, and Smarter Planet are trademarks or registered trademarks of IBM Corporation.

Linux is the registered trademark of Linus Torvalds in the United States and other countries.

Adobe, the Adobe logo, Acrobat, PostScript, and Reader are trademarks or registered trademarks of Adobe Systems Incorporated in the United States and other countries.

Oracle and Java are registered trademarks of Oracle and its affiliates.

UNIX, X/Open, OSF/1, and Motif are registered trademarks of the Open Group.

Citrix, ICA, Program Neighborhood, MetaFrame, WinFrame, VideoFrame, and MultiWin are trademarks or registered trademarks of Citrix Systems Inc.

HTML, XML, XHTML, and W3C are trademarks or registered trademarks of W3C®, World Wide Web Consortium, Massachusetts Institute of Technology.

Apple, App Store, iBooks, iPad, iPhone, iPhoto, iPod, iTunes, Multi-Touch, Objective-C, Retina, Safari, Siri, and Xcode are trademarks or registered trademarks of Apple Inc.

IOS is a registered trademark of Cisco Systems Inc.

RIM, BlackBerry, BBM, BlackBerry Curve, BlackBerry Bold, BlackBerry Pearl, BlackBerry Torch, BlackBerry Storm, BlackBerry Storm2, BlackBerry PlayBook, and BlackBerry App World are trademarks or registered trademarks of Research in Motion Limited.

© 2012 SAP AG. All rights reserved.

Google App Engine, Google Apps, Google Checkout, Google Data API, Google Maps, Google Mobile Ads, Google Mobile Updater, Google Mobile, Google Store, Google Sync, Google Updater, Google Voice, Google Mail, Gmail, YouTube, Dalvik and Android are trademarks or registered trademarks of Google Inc.

INTERMEC is a registered trademark of Intermec Technologies Corporation.

Wi-Fi is a registered trademark of Wi-Fi Alliance.

Bluetooth is a registered trademark of Bluetooth SIG Inc.

Motorola is a registered trademark of Motorola Trademark Holdings LLC.

Computop is a registered trademark of Computop Wirtschaftsinformatik GmbH.

SAP, R/3, SAP NetWeaver, Duet, PartnerEdge, ByDesign, SAP BusinessObjects Explorer, StreamWork, SAP HANA, and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP AG in Germany and other countries.

Business Objects and the Business Objects logo, BusinessObjects, Crystal Reports, Crystal Decisions, Web Intelligence, Xcelsius, and other Business Objects products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of Business Objects Software Ltd. Business Objects is an SAP company.

Sybase and Adaptive Server, iAnywhere, Sybase 365, SQL Anywhere, and other Sybase products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of Sybase Inc. Sybase is an SAP company.

Crossgate, m@gic EDDY, B2B 360°, and B2B 360° Services are registered trademarks of Crossgate AG in Germany and other countries. Crossgate is an SAP company.

All other product and service names mentioned are the trademarks of their respective companies. Data contained in this document serves informational purposes only. National product specifications may vary.

The information in this document is proprietary to SAP. No part of this document may be reproduced, copied, or transmitted in any form or for any purpose without the express prior written permission of SAP AG.