A NEW ARCHITECTURE FOR ENTERPRISE APPLICATION SOFTWARE BASED ON IN-MEMORY DATABASES

S

© 2015 SAP SE. ALL RI GHT RESERVED.

A NEW ARCHITECTURE

FOR ENTERPRISE APPLICATION

SOFTWARE BASED ON

IN-MEMORY DATABASES

Dr. Jens Krueger

SVP, Head of LoB Finance &

SAP Innovation Center

45

th

Annual Convention of the

German Informatics Society

September 30, 2015 Cottbus

S

2

Large enterprise systems

Transaction-maintained aggregates explode in complexity

~ 30,000 Functions

~ 80,000 Tables

~900,000 Columns

Data Footprint

of up to 150TB

}

Aggregate Maintenance

Development Costs

Schema Complexity

Consistency

Redundancy

Storage Costs

Backup

Disaster Recovery

}

}

S

Hardware development

Enabler for a new era of business processing

NEAR FUTURE

20 YEARS AGO

TODAY

MEMORY 48 TB CORES 480 (8X4X15) MEMORY

1

GB CPU

4

X

50

MHZ MEMORY

6

TB CPU

120

X

3.0

GHZ X 6000 X 1800 S

Workload analysis

Read-mostly applies to OLTP and OLAP

0 % 10 % 20 % 30 % 40 % 50 % 60 % 70 % 80 % 90 % 100 % OLTP OLAP W orkload 0 % 10 % 20 % 30 % 40 % 50 % 60 % 70 % 80 % 90 % 100 % TPC-C W orkload Read Insert Modification Delete Read Insert Modification Delete

TPC-C Benchmark _{Workload Analysis}

Krueger et al. VLDB’11

S

Simplification and Fexibility

By removing static pre-defined materialized aggregates

Static Pre-Defined

Aggregates Flexible AggregationOn Demand

Dynamic on-the-fly reporting enables

analysis & simulation of organizational changes. Impact analysis is

immediately available

FI POSTINGS CUSTOMERS PRODUCTS PROFIT CENTERS SUPPLIERS

ON-THE-FLY AGGREGATIONON-THE-FLY AGGREGATION

FI POSTINGS CUSTOMERS PRODUCTS PROFIT CENTERS SUPPLIERS

Static pre-defined aggregates cannot

handle structural changes within the organization

Requires Anticipation Does NOT Require Anticipation

S

2009: A common database approach

For OLTP and OLAP using an in-memory column database

Read-only Replicas Main Memory Storage

Data Management Layer OLTP & OLAP Applications

...

Logs _Checkpoints SQL Interface Stored Procedures Non-volatile Storage Stored Procedures Row Store Column Store Logistics Financials Manufacturing ... Dictionaries Attribute Vectors Aggregate Cache Attribute Vectors Dictionaries Aggregate Cache Attribute Vectors Dictionaries Aggregate Cache Attribute Vectors Dictionaries Main Main Main Delta

Actual Partition (Master)

Historical

Partition 1 Partition 2Historical

Source: SIGMOD June 2009

S

A textbook example

Wire money from account A to account B

Accounts with

Transaction-Maintained Balance

TRANSACTIONS BALANCE

Raw Account

Transactions

TRANSACTIONS Update Insert

Σ Read (Column Scan) Primary Key Select

S

Example

Posting of a vendor invoice

Vendor Invoice

Row 1: ~~~~~

Tax: ~~~~~

Total: ~~~~~

Header

Account

Vendor

Cost

G/L Account

Tax

G/L Account

119 Credit

100 Debit

19 Debit

S

SAP Financials – Invoice Posting

Before simplification

BSET

Index on Primary Key Other Indices

Application managed materialized views

Closed Items Tax Documents BKPF Accounting Document Header BSEG Accounting Line Items Application managed materialized aggregates COBK Controlling Object Document Header COEP Controlling Object Line Items

Vendors Customers G/L Accounts

BSAK BSIK BSAS BSIS BSID BSAD Open Items Update FAGLFLEXA Split Accounting Line Items Insert Customer KNA1 KNC1 Vendor LFC1 LFA1 General Ledger GLT0 SKA1 Cost Center COSP CSKS S

SAP Financials – Invoice Posting

Simplification

Customer KNA1 KNC1 BSET

Vendor General Ledger

LFC1 LFA1 SKA1 Tax Documents Cost Center COSP CSKS BKPF Accounting Document Header BSEG Accounting Line Items ACDOCA Split Accounting Line Items COBK Controlling Object Document Header COEP Controlling Object Line Items

Vendors Customers G/L Accounts BSAK BSIK BSAS BSIS BSID BSAD GLT0 SKA1 S

SAP Financials – Invoice Posting

After simplification with database views

Customer KNA1

Vendor General Ledger

LFA1 SKA1 Cost Center CSKS BKPF Accounting Document Header BSEG Accounting Line Items KNC1 LFC1 GLT0 COSP

Views with

on-the-fly

Aggregations

BSAK BSIK BSAS BSIS BSID BSAD

COBK COEP BSET

Views with

Projection and

Selection

ACDOCA Split Accounting Line Items S

Why column-stores are better than row stores for

transaction processing - theoretically...

Classic SAP Financials with Transaction-Maintained Aggregates

19 Inserts 5 Updates

Simplified SAP Financials

BSEG

BKPF BSEG BSIS BSIS BSIK BSET LFC1 GLT0 COBK COEP COSP

ACDOCA (PER LEDGER) FAGLFLEXA

(PER LEDGER) ...

BKPF

Insert duration with and w/o transaction-maintained aggregates

7 Inserts 0 Updates

S

...and experimentally with respect to workload and runtime

0 % 10 % 20 % 30 % 40 % 50 % 60 % 70 % 80 % 90 % 100 % Classic Simplified W orkload Insert Modification Delete

SAP Financials Workload Analysis SAP Financials Runtime Analysis

Document posting measured on customer data

SAP Financials Read 0 5000 10,000 15,000 20,000 25,000 30,000 Time (ms)

Suite on HANA S/4HANA Finance

Sum of CPU time (ms) Sum of DB reg. time (ms)

S

Universal Journey Entries establish a single source of truth

and enable a variety of new analyses

Analytics

Processing Transaktional Analytisch

Financial Accounting Documents Material Ledger Documents Asset Accounting Documents Dokumente der Finanzbuchhaltung Dokumente der

Finanzbuchhaltung Universal Journal

Entries Controlling Documents Profitability Documents Analytics Processing

Totals + Indices Totals + Indices Totals + Indices Totals + Indices Totals + Indices

on-the-fly aggregation

S

Compatibility views enable a non-disruptive adoption

by redundant tables

CORE DATA MODEL

OF SAP S/4HANA FINANCE

Filter Project Transform Filter Project BKPF ACDOCA Filter Project Aggregate Item Views

BSID, BSAD, BSIK, …, FAGLBSIS, … Aggregate Views GLT0, LFC1, KNC1, FAGLFLEXT, … Journal Views COBK, COEP, FAGLFLEXA COMPATIBILITY VIEWS BSEG S

No aggregates: on-the-fly aggregation of inventories

Simplified Logistics

Classical Logistics Data Model

Simplified Logistics Data Model

MSTQH MSKUH MSTBH MKOLH MSKAH MSTHE MSSAH MSEG

TRANSACTIONAL DATA AGGREGATED QUANTITIES MASTER DATA + AGGREGATED QUANTITIES MASTER DATA

MATDOC MKPF MARDH MCHBA MSSQH MCHBH MARCH MSTQ MSPR MSSQ MSKU MSLB MSTE MKOL MCHB MARD MSSA MSKA MARC MCH1 MAKT MCHA MARM MARA MSSQ MSKU MARD MSSA MSKA MARC MCH1 MAKT MCHA MARM MARA S

Simplified Logistics

Example case with real data from large automotive company

Backflush real-time, large automotive company

0 0 10 20 30 40 50 60 70 80 90 100 200 300 400 500 600 700 800 900 Delete Throughput items/sec # Parallel Processes Classical Data Model Simplified Data Model

S

Database footprint reduction

With SAP HANA

BACKUP/RECOVERY 1

/35

OF ERP ON ANY DB DEVELOPMENT SYSTEM 1

/35

_{OF FULL DATA} TEST SYSTEM 1

/35

_{OF FULL DATA} No Indices No Aggregates No Redundancy 7.1 TB 1.8 TB 0.8 TB

«sERP» on HANA (estimate) ERP on any DB ERP on HANA CACHE 0.3 TB CACHE 0.3TB HOT 0.2TB HOT 0.2TB 3 : 1

Hot = Actuals, status at beginning of current year + all transactions during

current year to conduct business and fulfill statuary reporting

Cold = Historical, up to 10 years

S

Transparent aggregate cache

Database-maintained aggregates on read-only main partition

MAIN COLD 1 COLD N DELTA MAIN COLD 1 COLD N DELTA AGGREGATES On-the-fly

Aggregation Aggregate CacheTransparent

Inserts

Aggregation Σ

Aggregation Σ

Inserts

S

Read-Only Replication

Leveraging business workload characteristics

Executives Sales Managers

Data Entry

Master Node

Read-only Replicas

Operational Reporting & New Applications

Accountants

OLAP, Search, and Read-only Applications on Transactional Schema Mixed Workload Processing OLTP Asynchronous, Transactionally Consistent Replication S

Boardroom Redefined

Review the state of the business at any time, examining data in real-time

Prioritize where we spend time reviewing Apply analytics to identify solutions

Higher speed leads to more intelligence

TRENDS

S

Thank you.

S

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 devel-opments 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

as-sumes no responsibility for errors or omissions in this document, except if such damages were caused by SAP intentionally or grossly negligent.

S

© 2015 SAP SE. ALL RI GHT RESERVED.