Highly Automated IT

Copyright © 2017, Oracle and/or its affiliates. All rights reserved.
Oracle Autonomous Database
and
Highly Automated Cyber Security
Larry Ellison
Chief Technology Officer
Robots
Fight Data Theft

Safe Harbor Statement
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,
timing and price of any features or functionality described for Oracle's products may change and remains at
the sole discretion of Oracle Corporation. Fees apply for new Database product offerings. 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. A detailed discussion of these
factors and other risks that affect our business is contained in our SEC filings, including our most recent
reports on Form 10-K and Form 10-Q, particularly under the heading "Risk Factors." Copies of these filings
are available online from the SEC or by contacting Oracle Corporation’s Investor Relations Department at
(650) 506-4073 or by clicking on SEC Filings on Oracle’s Investor Relations website at
http://www.oracle.com/investor. All information set forth in this presentation is current as of October 1st,
2017. Oracle undertakes no duty to update any statement in light of new information or future events.
2

• New Cyber Security Technology - Tuesday
– Automated Threat Detection and Immediate Remediation
• New Database Technology - Today
– Automated Database Can Instantly Patch Itself while Running
– No Delay Waiting for Human Process or Downtime Window
– Lots of Other Benefits from Database Automation
3
Automation to Fight Data Theft
Automated Cyber Security & Automated Database Work Together

Based on a Technology
As Revolutionary
As the Internet

Based on a Technology
As Revolutionary
As the Internet
Machine Learning

Machine Learning (ML)
• Autonomy: “Self-Driving” cars
• Computer Vision: Facial Recognition
• New ML Applications: Autonomous Database & Automated Cyber Security
The Applications are Revolutionary
6

Machine Learning
• Machine learning relies on large amounts of accurate training data
–Higher volumes of accurate data increases learning
–Increased learning/training means more accurate predictions
• Anomaly Detection: Separate normal from abnormal patterns in the data
–Detect a cancer cell from normal cells
–Detect the CFO logging on from a computer in the Ukraine
Computers Learn from Patterns in the Data and Make Predictions
7

Huge Amounts of Computer Systems Data in Event Logs
• Lots of Event Logs
– Infrastructure logs: Network, Server, Storage, VM, OS
– Platform logs: Database, Java, Analytics, etc.
– Applications logs: ERP, CX, HCM, Custom, etc.
• Event Log Training Data Enables New Machine Learning Applications
– Security: Detect and connect anomalous events: Login from Ukraine and unique SQL
– Database: Classify normal query patterns and automatically tune database
8
Enable New Database and Security Applications for Machine Learning

• Database Autonomy: Fully automated 100% “self-driving” database
• Automated Cyber Defense: Detect & block attacks in near real-time
• They Work Together:
–Discover attack: Real time ML log processing detects security anomaly in data
–Remediate: Database automatically patches itself while running
Lots of Other Benefits Come with Total Database Automation
9
Database Autonomy & Highly Automated Cyber Security

Announcing Oracle 18c
World’s First 100% Self-Driving
Autonomous Database

• No Human Labor: Eliminate human labor to manage the database
• Database automatically provisions, upgrades, patches, tunes itself while running
– Automated real-time security patching with no downtime window required
• No Human Error: SLA Guarantees 99.995% reliability and availability
– Minimize costly planned plus unplanned downtime to less than 30 minutes a year
• No Human Performance Tuning: Consumes less compute and storage than at Amazon
– Lower labor costs is an even bigger savings.
11
Oracle Autonomous Database 18c
Total Automation Based on Machine Learning

OLTP
– 4x faster in-memory OLTP access
– 5x faster RAC for high contention OLTP
– 2x faster streaming ingest for IoT workloads
– Zero impact grid infrastructure patching
– PDB cloud provisioning automation
– NVRAM ready row store, column store, SQL
Analytics
– 2x faster in-memory column store
– 100x faster In-memory analytics for external data
– 100x faster approximate query processing
– Query and Load directly from Oracle Object Store
– Active Data Guard for data warehouse loads
– New machine learning algorithms
12
Lots of New OLTP and Data Warehouses Features

• Fully Automated Database Provisioning and Management
– Even for mission critical scale-out clusters with datacenter disaster protection
– User defines policies then system automatically manages itself
– Automatic provisioning, backup, upgrades, patching, tuning, etc., while running
– No human administration means no administrator errors
13
No Human Labor – No Human Error

• Fully Automated Performance Tuning
– Automatic optimization: indexing, caching, statistics, parallelism, etc.
– Continuous tuning to adapt to data volume increases and data distribution changes
– Consumes less computer resources than human tuned system
14
Continuous Adaptive Performance Tuning

• Fully Automated Hardware Resource Elasticity
– Instantly expand/shrink compute and storage without downtime
– Minimizes cost by delaying resource allocation until time needed
– No overpaying for partially used fixed configurations
15
Dynamic Elastic Hardware Allocation

- System Failure or Scaling - Exadata, RAC, ASM
- Regional Outage or DB Corruption - Active Data Guard
- Patches (Updates) - RAC Rolling Upgrade
- Major Release Upgrades - Transient Logical Standby
- Table/Index/Schema Changes - Online and Edition Based Redefinition
- User Error - Flashback transaction, table, or database
16
Non-Stop Availability: Fault Tolerant plus On-line Recovery and Upgrades

• SLA Guarantees 99.995% Reliability and Availability
– Less than 30 minutes of downtime per year
– Includes both planned and unplanned downtime
– No exceptions in fine print: Up to 100x more reliable than Amazon
17
Guaranteed 99.995% Reliability and Availability

Autonomous Database for Data Warehouses
Available: December 2017
• Automated
–Automatically provision, upgrade, patch, tune while running
–Eliminates human labor and human error
• Efficient
–Minimize Hardware Resource Consumption
• Reliable
–Guaranteed 99.995% availability
–Up to 100x better availability than Amazon
18

Customer Data Center
Purchased
Customer Managed
Exadata Cloud at Customer and Public Cloud
On-Premises
Customer Data Center
Subscription
Oracle Managed
Cloud at Customer
Oracle Cloud
Subscription
Oracle Managed
Public Cloud Service
19

Introducing Database BYOL to PaaS Pricing
Includes Full Database Autonomy plus all Features Customer has Licensed
20
Oracle Public Cloud, Cloud@Customer BYOL PaaS Price
Storage
BYOL PaaS Price
Compute
Oracle Database Enterprise Edition
Standard Infrastructure
$50/Terabyte $168/OCPU
Oracle Database Enterprise Edition
Exadata Infrastructure
$148/Terabyte $280/OCPU
PaaS Discount of 30% with Monthly Dollar Volume Commitment
Minimum Configuration 1 OCPU plus 1 Terabyte: $300 per Month
BYOL EE Database Plus RAC, Multitenant, Active DataGuard for 99.995% Availability
Oracle Autonomous Database Runs on Exadata Infrastructure
Monthly Pay as you go Database PaaS Pricing

Oracle Autonomous Services for Every Database Workload
More Automation and Lower BYOL PaaS Prices: Middleware, Analytics…
• Autonomous Data Warehouse Cloud Service
• Autonomous OLTP Database Cloud Service
• Autonomous Express Database Cloud Service
• Autonomous NoSQL Database Cloud Service
21

Autonomous Database for OLTP
Available: June 2018
• Automated: upgrade, patch, tune… while running
– Handles OLTP and Mixed workloads
– Provision Mission Critical with High-Performance 99.995% availability configuration
– Provision Low Cost for non-critical workloads or test/dev with single server config
• Efficient: Consume fewer computer resources
– Running faster on Exadata infrastructure consumes less compute time
– Data Compression consumes less storage
– Instant scaling of compute or storage only when needed with no downtime
• Reliable: 99.995% Availability
22

Autonomous NoSQL Database Cloud Service: CY 2018
Sharded Key-Value for IoT, User Profiles, etc.
• Easy
– Fully-managed, configured for high availability/scalability
– Provision, load, and go
• Fast
– Predictable low latency (single digit millisecond) access
– Leverages cluster of next-gen IaaS dense IO servers
• Elastic
– Instant scaling of throughput or storage with no downtime
* Expected CY 2018
23

Database Professionals: Evolution of Skill Set
Problem: More data management tasks than humans to do the work
Less time on Administration
• Less time on infrastructure
• Less time on patching, upgrades
• Less time on ensuring availability
• Less time on tuning
More time on Innovation
• More time on database design
• More time on data analytics
• More time on data policies
• More time on securing data
24

SIX SHORT DEMOS
Oracle Autonomous Cloud Database
versus
Oracle Database on Amazon Cloud
and
Redshift Database on Amazon Cloud

Amazon’s Best Oracle Database Cloud Service is Called RDS
• Not Automated
–Manually upgrade, patch, tune…
–Expensive human labor, error prone human processes
–Requires repeated system downtime
• Not Efficient
–High Hardware Resource Consumption
–Consumes more than double the compute and storage
• Not Reliable
–Not Fault tolerant, no RAC
–No Data Guard for Regional Disaster Recovery
26

Amazon 99+% Reliability/Availability Guarantee Not Real
Amazon SLA excludes most sources of unplanned and planned downtime
• Amazon excludes downtime for adding compute and storage capacity
• Amazon excludes downtime for Amazon planned maintenance
• Amazon excludes downtime for database upgrades and patching
• Amazon excludes downtime for regional outages
• Amazon excludes downtime for software bugs
• And on and on and on….
https://aws.amazon.com/rds/sla/
27

Demo Configurations and Pricing
Real Customer Workloads on Same Size Hardware Configurations
• Oracle Autonomous Database Cloud Configuration
– Compute: 16 OCPU’s on Exadata x6 not Exadata x7
– Pricing: Standard BYOL PaaS pricing at 30% standard published discount
• Oracle Database on Amazon Cloud Configuration
– Compute: i3.16x large AWS’s largest high-IO server
– Pricing: Standard Amazon published pricing
• Amazon Redshift Database Configuration
– Compute: 8 x ds.xlarge nodes
– Pricing: Standard Amazon published pricing
28

Demonstration Workloads:
Real Data Warehouses from Four Different Customers
• Financial analytics
• Market research analytics
• Insurance analytics
• Retail analytics
29

DEMOs 1, 2, 3
Oracle Database on Oracle Cloud
versus
Oracle Database on Amazon Cloud

Demo #1:
versus
Oracle-on-AWS
• Financial Analysis Workload
31

Amazon
5.75x More Expensive
35

Demo #2: Oracle-on-AWS vs
Oracle Autonomous Database Cloud
• Retail Analysis Workload
36

Amazon
40

Demo #3: Oracle-on-AWS vs
• Insurance Analysis Workload
41

Amazon
8.3x More Expensive
45

• Efficient
–Oracle on Amazon 5 to 8 Times Higher Cost than Oracle Autonomous DW
–Save Even More on Reduced Labor Costs
46

DEMOs 4, 5, 6a-6b
Oracle Database on Oracle Cloud
versus
Amazon Redshift on Amazon Cloud

Demo #4: AWS Redshift vs
• Financial Analysis Workload
48

Amazon
9x More Expensive
52

Demo #5: AWS Redshift vs
• Retail Analysis Workload
53

Amazon
57

Demo #6a: AWS Redshift vs
Oracle Autonomous Database Cloud (8 OCPU)
• Market Research Analysis Workload
58

Amazon
62

Demo #6b: AWS Redshift vs
Oracle Autonomous Database Cloud (16 OCPU)
• Market Research Analysis Workload
63

Demo 6b: Dynamic Elastic Scale-Up While Running
• Scaling Oracle DWCS from 8 OCPUs to 16 OCPUs is Elastic
– Scaling occurs instantly
– Performance of workload nearly doubles
– Cost of workload is same
• Scaling Redshift from 8 nodes to 16 nodes is not Elastic
– Long-running operation: requires full unload – reload of database
– Requires downtime
• 8-node Redshift is converted to read-only
• Data copied to new 16-node Redshift
• 16-node Redshift is opened; 8-node Redshift is deleted
64

Instant Elasticity:
Grow to 16 OCPU online
65

Amazon
68

• Efficient
–Amazon Redshift 9 to 15 Times Higher Cost than Oracle Autonomous DW
–Save Even More on Reduced Labor Costs
69

Oracle Autonomous Database: Lower Labor Costs
ORACLE is Fully Automated – AWS Redshift is Manual
No tuning required
• Easy: define tables, load data, run
• Oracle automatically tunes:
• Automatic Compression
• Automatic Caching
• Automatic Indexing
Performance tuning required
• Customers must choose distribution
keys and sort key for each table
• Requires tuning expertise
No administration required Administration required
Oracle
Amazon
AWS
70

Oracle Autonomous Database Consumes Fewer Resources
Instant Resource Elasticity – AWS Redshift Not Elastic
Fully flexible: any combination of
compute and storage
Only available for fixed hardware
configurations
Instant scaling up or down of
compute and storage
• No downtime
Scaling a cluster requires a full
rebuild of database
• Downtime required
Oracle
Amazon
AWS
71

Safe Harbor Statement
The preceding 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,
timing and price of any features or functionality described for Oracle's products may change and remains at
the sole discretion of Oracle Corporation. Fees apply for new Database product offerings. 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. A detailed discussion of these
factors and other risks that affect our business is contained in our SEC filings, including our most recent
reports on Form 10-K and Form 10-Q, particularly under the heading "Risk Factors." Copies of these filings
are available online from the SEC or by contacting Oracle Corporation’s Investor Relations Department at
(650) 506-4073 or by clicking on SEC Filings on Oracle’s Investor Relations website at
http://www.oracle.com/investor. All information set forth in this presentation is current as of October 1st,
2017. Oracle undertakes no duty to update any statement in light of new information or future events.
72

Highly Automated IT

Recommended

More Related Content

What's hot (20)

Similar to Highly Automated IT (20)

More from Andrey Akulov (20)

Recently uploaded (20)

Highly Automated IT