Optimize Oracle Operations

 

Unload, Load, Reorg, Query, Mask and Test

Challenges

 

Oracle users face performance bottlenecks, data movement, and security issues like these:

  • unloading large tables
  • loading large tables
  • routine utility operations (reorgs)
  • complex queries or outer joins
  • data (warehouse) integration (ETL)
  • table migration or replication
  • data profiling, classification and E-R diagramming
  • data masking or encryption
  • SQL transforms or reports
  • smart test data creation or subsetting

 

Typical performance optimization solutions for Oracle involve complex query analyses, histograms, custom tuning, and/or  expensive appliances. Oracle column encryption can be cumbersome, and PII masking and test data capabilities are limited or typically available from overpriced megavendors, or specialty software providers who cannot address related governance requirements.

Solutions

 

Individual IRI software products (or the larger IRI Voracity platform) address these issues. IRI services are also available for custom requirements.

Start with Oracle and other source data discovery and classification in the free IRI Workbench GUI, which can also administer Oracle and many other databases at once.

Then, get ready to:

Use IRI FACT for Fast Extraction from Oracle. Dump data in parallel from very large Oracle tables to flat files in seconds. Flat files are the fastest,simples way to stage big data, and the only portable interchange format for large scale data integration, migration, and backup/restore.

Use IRI CoSort to speed Oracle loads:

  •    sort on the primary index key of the target table
  •    pipe the output to an SQL*Loader operation with the DIRECT=TRUE parameter
  •    rebuild the indexes after the load

Pre-sorted, direct path loads are the fastest method for bulk table population.

Use IRI Voracity as a faster, lower cost ETL alternative to the ELT operations of ODI and OWB.

Value lookups and joins are faster when data are in key order. Reorganize tables often to:

  •    maintain rows in the order of the most common, long-running queries
  •    recliam table space
  •    create indexes faster
  •    speed error-checking through views

The Offline Reorg wizard in the IRI Workbench (Eclipse GUI) automates the specification and use of FACT, CoSort, and SQL*Loader to perform en-masse Oracle table reorgs.

Use this wizard or any of its component pieces to optimize table reorderingw ithout impacting database users. With the space and time saved by offloading transforms and automating SQL*Loader, you can easily create and maintain multiple query tables.

Use the data source explorer and IRI NextForm or CoSort SortCL jobs in the IRI Workbench to select and subset the data you need to migrate and replicate. Discover sources and define metadata. Acquire, target, and map data in the GUI. Define new data stores and formats ad hoc or en masse. Account for format masks, keys, and other constraints.

Sensitive value discovery and column encryption using Oracle methods can be cumbersome to implement. Ramp-up data security and privacy law compliance faster -- with more options and auditability -- using IRI FieldShield (or Voracity which includes FieldShield). FieldShield offers a vast, database-agnostic array of easy-to-implement data discovery, classification, and column-level data masking functions like encryption, redaction, and pseudonymization. Call FieldShield standalone or from database applications to protect table data at rest or in motion through queries.

SQL transforms and reports perform order by, join, and group by processing inside the database. That is an inherently inefficient place to take those actions when data volumes are large.

Online users can see a degredation in query performance, as the database has to adjust its internal resources to do the heavy lifting of data transformation. The SQL procedure defining those jobs is also lengthy and complex.

The same jobs designed in IRI Voracity ETL workflows or written directly in the explicit 4GL of CoSort's SortCL run much faster outside the database! See this page if you use Oracle Data Integrator (ODI) or Oracle Warehouse Builder (OWB).

The IRI Workbench (Eclipse GUI) environment supports SortCL operations for single-pass transformation and reporting, or fast hand-offs to OBIEE and other BI tools (including direct ODA feeds to BIRT in Eclipse). Use FACT (to extract) and SortCL instead of, or alongside, Oracle for big data integration and reporting.

Free up Oracle for the storage and retrieval it was designed for ... free up the money you've been saving for Exadata.

To rapidly populate huge Oracle DBs or ETL operations with safe, realistic data, use IRI RowGen (or Voracity which includes RowGen). RowGen uses your data models to automatically generate and populate structurally and referentially-correct test data. There is no need to access or mask production data to achieve data realism. Alternatively, you can build production subsets automatically with the subsetting wizard in IRI Workbench, which will run with a RowGen, FieldShield, or Voracity license.

Working on-premise or in the Oracle Cloud!