Slide 1: Slide 1
Slide 2: Safe Harbor Statement
Safe Harbor Statement "Safe Harbor" Statement: Statements in this presentation relating to Oracle's future plans, expectations, beliefs, intentions and prospects are "forward-looking statements" and are subject to material risks and uncertainties. Many factors could affect our current expectations a
Slide 3: Product direction disclaimer
Any information regarding future product releases that is shared in this meeting is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any material, code, or functionality,
Slide 4: Oracle Exalytics
Oracle Exalytics Thomas Kurian Executive Vice President
Slide 5: Oracle Exalytics Summary
Oracle Exalytics Summary • In-Memory Analytics has made rapid advances – Memory is Faster, Cheaper, and has More Capacity Today • Oracle has the most mature & best In-Memory Technology – For Transaction Processing, Business Analytics, and Unstructured Information Processing • Oracle Exalytics is a c
Slide 6: Understanding In-Memory Computing
Understanding In-Memory Computing
Slide 7: Why In-Memory?
Why In-Memory? Memory is Faster, Cheaper, & has more Capacity Today In-Memory Capacity In-Memory Cost In-Memory Significantly Increased Decreased Faster 2002 256MB/DIMM 2002 $0.2/MB HDD 5ms response 2012 16GB/DIMM 2012 $0.009/MB D-RAM 100ns response 64X More Capacity 25X Cheaper 50,000X Faster Faste
Slide 8: In-Memory Database
In-Memory Database Requirements for Transaction Processing & Business Analytics Requirement For Business For Transaction Analytics Processing In-Memory Data Caching In-Memory Columnar Storage In-Memory Row & Column Compression In-Memory Indexes In-Memory Query Optimizer (Predictabili
Slide 9: In-Memory Business Analytics: Types of Analytics Problems
In-Memory Business Analytics 5 Types of Analytics Problems • Operational Reporting: Report on information in real time – Which customer orders will be affected by delays in my NYC warehouse today? • Relational Query & Analysis: Analyze information in real time – How many of these customers had late
Slide 10: In-Memory Business Analytics: Types of Analytics Problems
In-Memory Business Analytics 5 Types of Analytics Problems Operational Reporting Query & Analysis Multi-dimensional OLAP OLTP Operational OLTP Warehouse or OLTP OLAP Data Store Data Mart Unstructured Information Planning & Budgeting Discovery OLTP Planning Warehouse Unstructured System Analytics
Slide 11: Each Type of Problem has different Performance Requirements
In-Memory Business Analytics Each Type of Problem has different Performance Requirements Operational Reporting Query & Analysis Multi-dimensional OLAP 1. How fresh is the data in 1. How fresh is the data in my 1. How fresh is the data in my my ODS? warehouse or mart? OLAP Cubes? 2. How fast is my re
Slide 12: Each Problem has different Performance Requirements
In-Memory Business Analytics Each Problem has different Performance Requirements Operational Reporting Query & Analysis Multi-dimensional OLAP Fast Data Movement Fast Data Movement Fast, On-Line Cube Builds Fast Query Response Time Fast Query Response Fast Query & Aggregate Calculation Fast Aggregat
Slide 13: Oracle Exalytics – Product Overview
Oracle Exalytics – Product Overview
Slide 14: Oracle Exalytics In-Memory Machine
Oracle Exalytics In-Memory Machine Endeca Information Discovery Unique Features In-Memory Query & Analysis Hyperion Planning & Budgeting In-Memory Multi-Dimensional OLAP OBI In-Memory Analytics In-Memory Reporting In-Memory Planning & Budgeting Endeca In-Memory MDEX Server Runs Packaged Ap
Slide 15: Oracle Exalytics Hardware
Oracle Exalytics Hardware • Memory – 1 TB D-RAM • Compute – 4 Intel® Xeon® E7-4870, 40 cores total • Networking – 40 Gbps InfiniBand – 2 ports – 10 Gbps Ethernet – 2 ports – 8 Gbps FibreChannel – 2 ports – 1 Gbps Ethernet – 4 ports • Storage – 3.6 TB HDD Capacity • Operating System – Linux
Slide 16: Oracle BI Foundation
Oracle BI Foundation Powerful In-Memory Query & Analysis • In-Memory Optimizations – In-Memory R-OLAP, In-Memory Reporting – Latency Reduction: 8-10X faster response time – Scalability Improvements: 2-4X more users Scorecards Query & Analysis • Data Federation Capabilities Dashboards – Consistent ca
Slide 17: TimesTen In-Memory Database
TimesTen In-Memory Database Powerful In-Memory Transactional & Analytics DBMS • The Leading In-memory Database – Fully Persistent (Updates) & Highly Available – Proven in mission critical deployments – Used by 1000s of Companies TODAY • New Analytical Functions – Grouping Operators: CUBE, ROLLUP, GR
Slide 18: Essbase In-Memory M-OLAP
Essbase In-Memory M-OLAP Powerful In-Memory M-OLAP & Planning Engine • The Leading Multi-Dimensional OLAP Server – Fully Persistent (Updates) & Highly Available – Proven in mission critical deployments – Used by 1000s of Companies TODAY • Fast Data Feed – Online Trickle Feed of Data – Online Cube Me
Slide 19: Operational Reporting
Oracle Exalytics Operational Reporting Golden Gate & ETL Data Load Oracle BI TimesTen (ODS) OLTP DBMS Oracle Exalytics High Performance Operational Reporting Super Fast Data Refresh: Golden Gate for Transaction Replication Super Fast Query Performance: In-Memory Pre-Cached Queries, Results, View
Slide 20: Query & Analysis: In-Memory Data Marts
Oracle Exalytics Query & Analysis: In-Memory Data Marts ETL Data Load Oracle BI Times Ten (In-Memory Data Mart) OLTP DBMS Oracle Exalytics High Performance Query and Analysis for Data Marts Identifying Hot Data to Cache in Mart: Oracle BI Summary Advisory High Capacity In-Memory Storage: Columna
Slide 21: Query & Analysis: Enterprise Warehouse with Oracle Exadata
Oracle Exalytics Query & Analysis: Enterprise Warehouse with Oracle Exadata Superfast Infiniband Oracle BI Times Ten Data Warehouse (Hot Data Cache) (Oracle Exadata) Oracle Exalytics High Performance Query and Analysis for Data Warehouses Fast Query and Analysis: Automatically move ‘Hot’ Data into
Slide 22: Multi-Dimensional OLAP
Oracle Exalytics Multi-Dimensional OLAP ETL Data Load Oracle BI 1 or More Data Sources Essbase Oracle Exalytics High Performance Multi-Dimensional OLAP On-line, Rapid Cube Building: Essbase In-Memory Fast Cube Rebuild and Aggregation: Fast Writes/Updates Scalable Forecasting and What-if Analys
Slide 23: Planning & Budgeting
Oracle Exalytics Planning & Budgeting ETL Data Load Hyperion EPM Applications Essbase ERP & Operational Applications Oracle Exalytics High Performance Planning & Budgeting Fast Plan Updates & Incremental Aggregation: Fast Block Writes Broader Scenario Modeling and Better Forecasting: Non-Layered
Slide 24: Unstructured Information Discovery
Oracle Exalytics Unstructured Information Discovery Unstructured Information Add Structure to Unstructured Data Oracle Endeca Information Discovery + Structured & Semi-Structured Oracle Endeca Server Oracle Exalytics Interactive Discovery on Unstructured Information Rapid Ingestion of Unstructured
Slide 25: Packaged Applications Support
Oracle Exalytics Packaged Applications Support Packaged Business Intelligence Applications Comms Complex Consumer Financial High Insurance Life Public Travel Auto Mfg Sector Energy Services Tech & Health Sciences Sector & Trans & Media Service & Order Human Sales Marketing Supply Chain Financials Co
Slide 26: Customer Results
Oracle Exalytics Customer Results • Largest mortgage provider in Denmark, major private bond issuer in Europe • 1700 Power Analytics Users; 50 Terabytes Data; Superfast Adhoc Query Performance • 35X to 70X faster with Exadata + Exalytics • Supplies automotive industry with market intelligence “PolkI
Slide 27: Competitive Comparisons
Competitive Comparisons
Slide 28: What are limitations of pure In-Memory DBMS?
What are limitations of pure In-Memory DBMS? In-Memory is NOT a panacea for all DBMS • Transaction Processing – Updates/Writes need to be stored on disk for High Availability or Durability – If Writes done to Disk, Sophisticated Clustering required to share data across Nodes • Data Warehousing (Scal
Slide 29: TimesTen is significantly better than SAP HANA
Oracle Exalytics Oracle TimesTen is significantly better than SAP HANA Requirement Oracle Times Ten SAP HANA In-Memory Data Caching In-Memory Columnar Storage In-Memory Row & Column Compression Column Only In-Memory Indexes In-Memory Query Optimizer (Predictability) In-Memory NUMA Supp
Slide 30: Exalytics is significantly better than SAP HANA
Oracle Exalytics Oracle Exalytics is significantly better than SAP HANA Exalytics SAP HANA Hana’s Limitations Operational Reporting Limited Data Sources with Sybase Replication Server Limited support of 3rd normal form within Business Objects Data Mart No Parallel Query (Scale-Out) or NUMA (Scale-Up
Slide 31: Exalytics is significantly cheaper than SAP HANA
Oracle Exalytics Oracle Exalytics is significantly cheaper than SAP HANA SAP HANA + Oracle Exalytics Comparison IBM Hardware In-Memory Data Mart Hana is 5X more 512GB compressed data $825,000 $4,112,474 expensive 1TB memory In-Memory Analytics for $126,548,960 Enterprise DW $2,499,000 Hana is 50X 40
Slide 32: Summary
Summary
Slide 33: Oracle Exalytics Summary
Oracle Exalytics Summary • In-Memory Analytics has made rapid advances – Memory is Faster, Cheaper, and has More Capacity Today • Oracle has the most mature & best In-Memory Technology – For Transaction Processing, Business Analytics, and Unstructured Information Processing • Oracle Exalytics is a c
Slide 34: Slide 34