Weitere ähnliche Inhalte Ähnlich wie Oracle Autonomous Database - introducción técnica y hands on lab (20) Kürzlich hochgeladen (20) 1. Hands-on lab
Oracle Autonomous Database
Copyright © 2020, Oracle and/or its affiliates | Confidential: Internal/Restricted/Highly Restricted1
Diego “Perico” Sanchez
Master Principal Sales Consultant
Oracle Argentina – Public Sector
diego.sanchez@oracle.com
/in/pericosanchez
Julio 2020
2. Declaración de puerto seguro - Información específica para España y Latinoamérica
La información recopilada por Oracle e incluida en el siguiente documento ha sido recopilada a través de un analista externo del sector y con el único fin de
obtener datos de interés sobre la dirección general de nuestros productos. Estos datos tienen fines puramente informativos, y no deben utilizarse para
diferenciar los productos de Oracle de los de otras empresas.
Oracle ha llevado a cabo una prueba conceptual con el más alto nivel de minuciosidad acorde al estándar del sector, pero no se garantiza que la información
del siguiente texto carezca de errores. Oracle no declara ni garantiza que los resultados de la prueba sean reproducibles o que puedan probarse lógicamente.
Oracle no proporcionará ningún dato relacionado con la prueba, entre otros: el método, el razonamiento, la idea, la estrategia seguida en la prueba o la
política de la prueba.
El propósito del siguiente texto es determinar la dirección general de nuestros productos. El presente documento ha sido redactado con fines informativos
únicamente y no se podrá incorporar a ningún contrato. Además, no supondrá compromiso alguno de entrega de ningún material, código o funcionalidad,
ni deberá utilizarse como base para la adopción de decisiones de compra. El desarrollo, lanzamiento, precio y plazo de puesta a disposición de cualesquiera
funciones o funcionalidades descritas para productos de Oracle podría cambiar y quedará a la sola discreción de Oracle Corporation.
Este material es la traducción de una presentación redactada originalmente en inglés para la sede de Oracle en Estados Unidos. La traducción se ofrece a
modo de cortesía y puede contener errores. Es posible que no todas las funciones y funcionalidades estén disponibles en todos los países ni en todos los
idiomas. Si tiene alguna duda, póngase en contacto con su representante de ventas de Oracle.
Declaración de puerto seguro - General
2 Copyright © 2020, Oracle y/o sus filiales. Todos los derechos reservados.
3. Copyright © 2020 Oracle and/or its affiliates.
5
4
3
2
1
Hands-on lab
Break
Demo
Under the hood
Introduction to Autonomous Database
Agenda
4. Copyright © 2019 Oracle and/or its affiliates.
Autonomous Database
Introduction
5. Copyright © 2020 Oracle and/or its affiliates.
Pionero en tecnologías de Internet, estratega y periodista
Creador del término “Podcasting” en 2004
Ben Hammersley
“Hace 5 años los autos sin
conductor eran ciencia ficción, hoy
son una realidad”
6. Oracle Spent Last 20 Years
Automating Database Technology
Copyright © 2019 Oracle and/or its affiliates.
• Automatic Query Rewrite
• Automatic Undo Management
• Autonomous Health Framework
• Automatic Diagnostic Framework
• Automatic Refresh of Clones
• Automatic SQL Tuning
• Automatic Workload Capture/Replay
• Automatic SQL Plan Management
• Automatic Capture of SQL Monitor
• Automatic Data Optimization
• Automatic Memory Management
• Automatic Segment Space Mgmt
• Automatic Statistics Gathering
• Automatic Storage Management
• Automatic Workload Repository
• Automatic Diagnostic Monitor
• Automatic Columnar Flash
• Automatic IM population
• Automatic Application Continuity
9i
10g
11g
12c
18c
• Automatic Indexes
• SQL Quarantine
• Real-Time Statistics 19c
7. Oracle Spent Last 10 Years Automating
Database Infrastructure
Copyright © 2019 Oracle and/or its affiliates.
• Smart Scan
• Infiniband Scale-Out
• Database Aware Flash
Cache
• Storage Indexes
• Hybrid Columnar Data
• IO Priorities
• Data Mining Offload
• Offload Decryption
• Direct-to-wire Protocol
• JSON and XML offload
• Instant failure detection
• Network Resource Mgmt
• Prioritized File Recovery
• In-Memory Columnar in Flash
• Smart Fusion Block Transfer
• Exadata Cloud Service
2008
2019
8. Eliminates All the Complexity of Mission Critical Databases
Autonomous Database Completes the Job
Copyright © 2019 Oracle and/or its affiliates.
Automated
Data Center
Operations and
Machine Learning
Complete
Infrastructure
Automation
Complete
Database
Automation
ORACLE
AUTONOMOUS
DATABASE
Oracle
Cloud
9. Autonomous Database Key Attributes
Self-Driving
Automates all database and
infrastructure management,
monitoring, tuning
Self-Securing
Protects from both
external attacks and
malicious internal users
Self-Repairing
Protects from all
downtime including
planned maintenance
Spend Less, Reduce Risk, Innovate More
Copyright © 2019 Oracle and/or its affiliates.
10. Traditionally DBAs are Responsible for:
10
• Tasks Specific to Business and Innovation
– Architecture, planning, data modeling
– Data security and lifecycle management
– Application related tuning
– End-to-End service level management
• Maintenance Tasks
– Configuration and tuning of systems, network, storage
– Database provisioning, patching
– Database backups, H/A, disaster recovery
– Database optimization
Value Scale
Innovation
Maintenance
11. • Maintenance Tasks
– Configuration and tuning of systems, network, storage
– Database provisioning, patching
– Database backups, H/A, disaster recovery
– Database optimization
11
Autonomous Database Removes Generic Tasks
• Tasks Specific to Business and Innovation
– Architecture, planning, data modeling
– Data security and lifecycle management
– Application related tuning
– End-to-End service level management
Value Scale
Innovation
Maintenance
12. One Autonomous Database – Two Deployment Choices
ORACLE
AUTONOMOUS
DATABASE
DedicatedServerless
Ultra-Simple and Elastic
Customizable Private Database
Cloud in Public Cloud
Copyright © 2019 Oracle and/or its affiliates.
13. Autonomous Database Serverless – Primary Benefits
• Simple
– Oracle automates and manages everything
– Customer just chooses database compute, storage, and region
• Elastic
– Low minimum size - 1 OCPU
– Low minimum time commitment - 1 hour
– Automatically scales online for true pay-per-use
Copyright © 2019 Oracle and/or its affiliates.
14. Autonomous Database Dedicated – Primary Benefits
• Provides your own Database Cloud running on
dedicated Exadata Infrastructure
• Highest Isolation
• Customizable Operational Policies
Copyright © 2019 Oracle and/or its affiliates.
15. One Autonomous Database – Optimized by Workload
ORACLE
AUTONOMOUS
DATABASE
Autonomous Transaction
Processing - ATP
Autonomous Data
Warehouse - ADW
Best for all Analytic Workloads:
• Data Warehouse, Data Mart
• Data Lake, Machine Learning
Best for TP and Mixed Workloads:
• Transactions, Batch, Reporting
• IoT, Machine Learning
Copyright © 2019 Oracle and/or its affiliates.
17. • APEX & ORDS
• SQL Dev-web
• Developer workbench and modeling
• Performance Hub SQL analysis
• Service SDKs
• Java, Python, Node, Go, .NET, Ruby
• See Developer Guide for TAC setup
17
Developer ready with APEX & ORDS
Currently, unique to Dedicated
19. Auto Indexing On Netsuite
• Ran a complex Netsuite workload, and compared Auto Indexing to tuned system
• 17,542 SQL statements, 1,852 tables, 8,151 indexes - years of tuning to create these indexes
• Before running on Auto Indexing, all indexes were dropped
Auto Indexing achieved near-identical performance to manual tuning
Auto Index DB stays tuned as workload changes
Elapsed
Time
(sec)
Manually
tuned
Auto Index
11551172
Indexes
Used
Manuall
y tuned
Auto Index
1733
4663
Copyright © 2019 Oracle and/or its affiliates.
20. Auto Indexing On Accounts Receivable
• Workload: 4,889 SQL statements running against 35 tables
• Indexes:
• Application code created 49 indexes of which 17 were used
• Automatic indexing created 5 indexes, all of which were used
Elapsed
Time
(sec)
Manually
tuned
Auto Index
4
123
Indexes
Used
Manuall
y tuned 5
17
Auto Index
23. Autonomous Database Platform
• RAC on Exadata
• Autonomous Database decides
where to place each database during
provisioning
• Fewer number of instances preferred
when possible
• Databases may be open on one
node
• Still RAC enabled
Copyright © 2019 Oracle and/or its affiliates.
24. Database Parameters
• All parameters set to optimal values based on workload type
• May be different than regular database defaults
• Users can only change a limited number of parameters
Copyright © 2019 Oracle and/or its affiliates.
Allowed Parameters
APPROX_FOR_AGGREGATION
APPROX_FOR_COUNT_DISTINCT
APPROX_FOR_PERCENTILE
FIXED_DATE
MAX_IDLE_TIME
NLS_CALENDAR
NLS_COMP
NLS_CURRENCY
NLS_DATE_FORMAT
NLS_DATE_LANGUAGE
NLS_DUAL_CURRENCY
NLS_ISO_CURRENCY
NLS_LANGUAGE
NLS_LENGTH_SEMANTICS
NLS_NCHAR_CONV_EXCP
NLS_NUMERIC_CHARACTERS
NLS_SORT
NLS_TERRITORY
NLS_TIMESTAMP_FORMAT
NLS_TIMESTAMP_TZ_FORMAT
OPTIMIZER_CAPTURE_SQL_PLAN_BASE
LINES
OPTIMIZER_IGNORE_HINTS
OPTIMIZER_IGNORE_PARALLEL_HIN
TS
PLSCOPE_SETTINGS
PLSQL_CCFLAGS
PLSQL_DEBUG
PLSQL_OPTIMIZE_LEVEL
PLSQL_WARNINGS
STATISTICS_LEVEL
TIME_ZONE
25. Optimizer Hints
• Optimizer and PARALLEL hints work differently based on
workload type
• Users can override by changing two parameters
Copyright © 2019 Oracle and/or its affiliates.
TRUE FALSE
optimizer_ignore_hints Ignores optimizer hints Honors optimizer hints
optimizer_ignore_parallel_hint
s
Ignores PARALLEL
hints
Honors PARALLEL hints
ADW Ignores optimizer and PARALLEL hints
ATP Honors optimizer and PARALLEL hints
26. Optimizer Statistics
• Stats are gathered automatically for direct load operations
• create table … as select …;
• insert /*+ append */ into … select …;
• Parallel inserts with or without the append hint
• Data Pump Import loads
• dbms_cloud loads
Copyright © 2019 Oracle and/or its affiliates.
27. Optimizer Statistics in 19c
• Real-time Statistics Collection
• Gathers a subset of optimizer statistics for conventional DML operations
• Number of rows, MAX and MIN column values, etc.
• High-frequency Statistics Collection
• Gathers full optimizer statistics every 15 minutes if statistics are stale
Copyright © 2019 Oracle and/or its affiliates.
29. User Management
• Pre-defined ADMIN user for database management
• Subset of SYSDBA privileges
• No SYS, SYSTEM access for users
Copyright © 2019 Oracle and/or its affiliates.
30. User Management
• Pre-defined role, DWROLE, for DW developers
• Includes common developer privileges
Copyright © 2019 Oracle and/or its affiliates.
DWROLE Privileges
CREATE ANALYTIC VIEW
CREATE TABLE
CREATE HIERARCHY
CREATE JOB
ALTER SESSION
CREATE ATTRIBUTE DIMENSION
CREATE TYPE
CREATE VIEW
CREATE PROCEDURE
CREATE SESSION
CREATE SEQUENCE
CREATE MINING MODEL
CREATE TRIGGER
CREATE SYNONYM
grant dwrole to sales identified by WelcomeSalesADW19;
32. Storage Management
• Provisioned storage size is net usable space with 3x redundancy
• Includes system tablespaces and the local filesystem
• 10% overage allowed
• Exadata flash size is proportional to the storage size
Copyright © 2019 Oracle and/or its affiliates.
Set storage size at provisioning time Scale up/down any time without downtime
33. Tablespace Management
• All tablespaces created and managed automatically
• Tablespace sizes depend on the provisioned storage size
• All tablespaces auto extend when needed
• Users cannot create or drop tablespaces
• All objects are created in DATA automatically
• Multiple DATA tablespaces created when storage size exceeds 32 TB
Copyright © 2019 Oracle and/or its affiliates.
Tablespaces from a Sample Autonomous Database
SYSTEM
SYSAUX
UNDOTBS1
TEMP
DATA
DBFS_DATA
SAMPLESCHEMA
UNDO_8
34. Table/Index Storage Attributes
• No need to worry about storage attributes
• Optimized and managed by Autonomous
Database
• Table/index storage attributes ignored
• Tablespace, INITIAL, NEXT, PCTFREE, PCTUSE,
etc.
Copyright © 2019 Oracle and/or its affiliates.
35. Compression
• Compression configured automatically based on ADB type
• Users can override using create/alter table commands, all
compression types available
Copyright © 2019 Oracle and/or its affiliates.
ADW Hybrid Columnar Compression, Query
High
ATP No compression
create table orders_history … column store compress for archive high;
37. Resource Allocation
• System resources allocated based on number of OCPUs
dynamically
• CPU, IO, number of sessions, SGA, PGA, etc.
• More OCPUs provide more resources
Copyright © 2019 Oracle and/or its affiliates.
Set CPU core count at provisioning time Scale up/down any time without downtime
38. Auto Scaling
• Enables the database to use up to 3x CPU/IO resources
immediately when needed by the workload
• Pay for exactly what you use
• Helps CPU or IO bound workloads
• Does not scale up other resources. Provision more OCPUs to
scale up these.
• Number of sessions
• Concurrency
• PGA, SGA, etc.
Copyright © 2019 Oracle and/or its affiliates.
39. SERVICE
DEFAULT SQL
PARALLELISM
SHARE OF
RESOURCES
CONCURRENT NO
OF QUERIES/DML
TPURGENT MANUAL 12 100 X OCPUs
TP 1 8 100 X OCPUs
HIGH OCPUs 4 3
MEDIUM 4 2 1.25 X OCPUs
LOW 1 1 100 X OCPUs
OLTP
DW, Batch,
Reporting
Pre-Defined Database Services
• Applications connect to a pre-defined database service
• e.g. jdbc:oracle:thin:@dbname_high?TNS_ADMIN=/Users/test/wallet_dbname
• Database services control resource usage
• SQL parallelism, CPU/IO shares, max concurrent queries/DML
Copyright © 2019 Oracle and/or its affiliates.
40. Which Service to Use
• OLTP
• Connect to TP or TPURGENT based on priority requirements
• DW
• Connect to HIGH, MEDIUM, LOW based on concurrency requirements
Copyright © 2019 Oracle and/or its affiliates.
HIGH
• Low concurrency
• More resources
per user
MEDIUM
• Higher concurrency
• Fewer resources per
user
LOW
• Highest concurrency
• Fewest resources per
user
42. Service Console
• Simple monitoring graphs for key service metrics
• Accessible by database users without a cloud account
Copyright © 2019 Oracle and/or its affiliates.
43. Performance Hub
• ASH Analytics and SQL Monitoring
• Accessible by users with cloud accounts
Copyright © 2019 Oracle and/or its affiliates.
46. ¡ Muchas gracias por
participar de este
hands-on !
Copyright © 2020, Oracle y/o sus filiales. Todos los derechos reservados.46
Diego “Perico” Sanchez
Master Principal Sales Consultant
Oracle Argentina – Public Sector
diego.sanchez@oracle.com
/in/pericosanchez
