Preview of Oracle Database 12c In-Memory Option. Copyright 2013, Oracle and/or its affiliates. All rights reserved.

(1)

Preview of

Oracle Database 12c In-Memory Option

(2)

The following 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, and should not be relied upon in making purchasing decisions. The development, release, and timing of any features or functionality described for Oracle’s products remains at the sole discretion of Oracle.

(3)

Oracle Database In-Memory Option Goals

100X Faster Queries: Real-Time Analytics

 Get instantaneous query results

 Querying OLTP database or data warehouse 2X Faster Transaction Processing

Trivial to Deploy for All Applications and Customers

(4)

Until Now You Choose One Format and Suffer Tradeoffs

Optimizing Transaction and Query Performance

Row Format Databases versus Column Format Databases

Row

 Transactions run faster on row format

– Fast for processing few rows, many columns

– Example: Insert or query a sales order

Column

 Analytics run faster on column format

– Fast for processing few columns, many rows

– Example: Report on sales totals by state

ORDER

SALES

S T A T E

(5)

 BOTH row and column in-memory formats for same table

 Simultaneously active and transactionally consistent

 Analytics & reporting use New Column format

 OLTP uses row format

Breakthrough: Dual Format In-Memory Database

Column Format Memory

Row Format

Memory

Analytics OLTP ^Sales ^Sales

(6)

Oracle In-Memory Columnar Technology

Memory

Pure Columnar

 Pure in-memory format with no logging

 Near zero overhead on changes

 Even for OLTP

 Uses memory-optimized compression

 2x to 10x memory reduction

 Data loaded in-memory for active tables or partitions - on startup or first access

 For in-memory tables, >90% of memory will be used for column format

 Row format needs little memory

(7)

Find Any Business Data in Sub-second

SIMD

Compare all values in 1 cycle

Compare all values

in 1 instruction Load

multiple State

values Vector Register In-Memory Column Store

State column

Sales

Example: Find all sales in state of CA

“CA”

>100X Faster

• Each CPU scans local in-memory columns

 Scans use super fast SIMD vector instructions

 Billions of rows/sec scan rate per CPU core

CPU

(8)

Scans and Combines Data from Multiple Tables

Sales

Stores

Type=outlet

Example: Find all sales in outlet stores

T Y P E

Storeid in 15,38,64

S T O R E I D

A M O U N T

 Converts join processing into fast column scans

 Joins up to 10x faster

Sum

(9)

Generate Reports In Sub-second

In-Memory Report Outline

Example: Show sales trends for footwear products in outlet stores

Stores

Products

Sales

 Dynamically creates in-memory report outline

 Report contents populated during fact scan

 Reports up to 20x faster without predefined cubes

(10)

OLTP is Slowed Down by Analytic Indexes

Table

1 to 3 OLTP Indexes

5 to 15 Analytics

Indexes

 Most Indexes in mixed-use OLTP (e.g. ERP) databases are only used for analytics

 Indexes work well for known access patterns both

in-memory and on-disk

 But every change to the table requires changing all analytic indexes – Slow!

(11)

Column Store Replaces Analytic Indexes

Table

1 to 3 OLTP Indexes

 In-Memory Column Store replaces analytic indexes for tables that fit in memory

 Removes analytic index overhead on changes

 Both predefined and ad-hoc analytic queries run fast

 Less tuning & admin needed

 OLTP & batch often run 2x or more faster

In-Memory Column Store

(12)

Demonstration at Oracle Openworld 2013

Performance of columnar scan vs. row scan

- Both in memory

(13)

Extreme Capacity and Cost Effectiveness

 Size not limited by memory

 Data transparently

accessible across tiers

 Each tier has specialized algorithms & compression

Capacity of Disk IOs of Flash

Speed of DRAM ^DISK

PCI FLASH DRAM

Cold Data Hottest Data

Active Data

(14)

Scale-Out In-Memory Database to Any Size

 Scale-Out across servers to grow memory and CPUs

 In-Memory queries are

parallelized across servers to access local column data

 Direct-to-wire InfiniBand protocol speeds messaging

In Memory Column Store

In Memory Column Store In Memory

Column Store In Memory Column Store

(15)

Scale-Up for Maximum In-Memory Performance

 Scale-Up on large SMPs

 SMP scaling removes overhead of distributing queries across servers or coordinating transactions

 Inter-processor bandwidth far exceeds any network

(16)

Oracle In-Memory is Trivial to Deploy

1. Configure Memory Capacity

 inmemory_area = XXXX GB

2. Configure tables or partitions to be in memory

 alter table | partition … inmemory;

3. Later Drop analytic indexes to speed up OLTP

(17)

Oracle In-Memory is Transparent to Applications

Existing Applications Just Run Faster

Full Functionality - No restrictions on SQL

Trivial to Implement - No migration of data or change of product Fully Compatible - All existing applications run unchanged DB as a Service Ready - Oracle Multitenant in-memory

Uniquely Achieves All In-Memory Benefits With No Application Changes

And All Other Apps that Support Oracle Database

(18)

Oracle In-Memory Delivers Extreme Availability

 Pure In-Memory format does not change Oracle’s storage format, logging, backup, recovery, etc.

 All Oracle’s mature availability technologies work transparently

 Protection from all failures

 Node, site, corruption, human error, change, etc.

RAC

ASM

RMAN

Data Guard & GoldenGate

(19)

• 100X Faster Queries: Real-Time Analytics

Querying OLTP database or data warehouse

• 2X Increase Transaction Processing Rates

• Less Management and Tuning

• Best of Memory, Flash, Disk

• Scale-Out and Scale-Up

• Extreme Availability

Trivial to Deploy for All Applications and Customers

Summary: Oracle Database In-Memory Option

(20)

APPENDIX

Relation of Database In-Memory Option to TimesTen and Exadata

(21)

Oracle Database In-Memory Products

Application

TimesTen In-Memory Database

 Embedded in custom applications

 Extreme low latency OLTP workloads (Microsecond Response Time)

 Adaptive In-Memory Cache for Exalytics In-Memory Machine

Oracle Database In-Memory

 Scalable In-Memory Database for any type of application workload

 Combines best of DRAM, flash, disk

(22)

TimesTen In-Memory Database and Columnar

 TimesTen is adding scale-out capabilities

– For both OLTP and analytics

 TimesTen is adding columnar technologies to accelerate Analytics

– Core columnar processing engine is shared between Oracle and Timesten

– Embedded Analytics engine

TimesTen Grid

Scale-Out In-Memory DB

(23)

Oracle Database In-Memory Option on Exadata

• Exadata extends fast analytics from in-memory to flash and disk for highest capacity and lowest cost

• Scale-out intelligent flash optimized storage: >100GB/sec from flash

• Exadata DB as a Service In-Memory delivers consolidation, ultra fast performance, best agility, and lowest cost

• Highest performance applications run in-memory

• Applications that perform well today use cost-effective disk and flash

• Expect special optimizations for Database In-Memory on Exadata

• E.g. Fastest communication with InfiniBand Direct to Wire protocol

• Exadata is a standardized, pre-configured, pre-tuned, pre-tested platform for database in memory, on flash, and on disk

(24)

Platform of the Future

 Platform Engineered for Database

 Database Engineered for Cloud and Consolidation

 Optimized for latest technologies

 In-Memory Database

 PCI Flash for fastest IO

 Disk for highest capacity

 Optimal compression at every tier

 Fastest Communication

 For all workloads – running together

 Highest performance, lowest cost, most available, most scalable

Enterprise DBaaS

(25)

(26)