In the 4.2.0.2.0 release of Oracle Utilities Application Framework, a new ILM based data management solution was implemented to allow customers to manage the increasing cost of storage for their Oracle Utilities products. It is being progressively rolled out across each of the Oracle Utilities products over the next year. With utilities across the world having to deal with larger and larger data volumes over time, the cost of storing that data and ensuring appropriate access by the business to that data become key concerns. ILM offers tools to address that.
The benefits of using this facility are as follows:
- ILM is about designing the lifecycle of your transaction data in a storage point of view. The first step of implementing the approach is for the business to define the data retention period for individual objects within the product itself. This defines the active period of the data, in days, where the business wants active update actions on the data. One of the major issues in data storage is the ability for the business and IT to commit to communicate when the business needs active access to the data. This addresses this by making it explicit in configuration.
- When a transactional object, that is ILM enabled, is created, the data is then automatically tracked internally using ILM specific fields. This defines when the date to consider the data active from (ILM_DT) and a flag indicating the ILM status of the record (ILM_ARCH_SW) in terms of the business. An easy way to remember this is that the date set for ILM is basically a ticking clock. As soon as the record is created, the ILM clock starts. It tells the product when to consider the before mentioned retention period from in respect to that individual object.
- As the ILM data is part of the object itself, it can be manipulated by any custom business process if desired. For example, say some customer transaction data was due to expire but the utility receives a complaint at the last minute on that data. It is possible to create a customization to delay the individual records, to keep them active, till the complaint is resolved. This basically means the ILM data can be manipulated by a business process, if desired.
- A dedicated background process, known as an ILM Crawler, assesses data when it expires to make sure nothing is outstanding on the object and indicates this state using an ILM flag. This tells the IT group that the individual data is now considered less-active by the business so the IT group can now manage the storage using Oracle's ILM facilities.
- The ILM flag effectively protects the data from any storage based solutions as long as it is needed by the business. The business can defer the lifecycle of the individual objects as needed and internal processes affecting the flag will ensure the data is protected if it is still active.
- The IT Group is free to reuse their storage solutions capabilities or use the ILM features within the database such as partitioning, compression and now in Oracle Database 12c, Automatic Data Optimization.
- The cost savings in terms of storage can be realized in a number of ways:
- Partitioning - Product tables can be partitioning along data retention times to realize cost savings using tiered storage solutions. For example, placing less active data on lower cost storage saves costs. Partitioning also means you can use Transportable Tablespaces to quickly remove data that is dormant (not needed by the business at all). Transportable tablespaces are also useful for restoring data.
- Compression - Oracle offers a full range of compression options from basic compression built into the database, the optional Advanced Compression option which offers flexible row level compression optimized for OLTP systems to the high performance HCC compression offered by Oracle ExaData,
- Automatic Data Optimization (ADO)/Heat Map - This allows real time tracking of data usage regardless of the state or business activity of the data. This feature coupled with Partitioning and/or Compression allows sites to realize additional savings regardless of the state data as it is based upon real usage rather than expected usage. For example, customers may want to keep two years of data as active on the database. In reality, they may not update the data after 1 year, ADO determines the last date/time the record was updated and can compress the data automatically if it detects that it has not been updated for a while, which realizes more savings.
- The tools for the technical definition of ILM are available from the command line, Oracle Database Control, Oracle EM Express, ILM Assistant or Oracle Enterprise Manager. These are tools that most DBA's and technical people will already be familiar with so skills can be reused in this respect.
The ILM based data management solution for Oracle Utilities, a powerful and flexible solution that marries the business needs for data against the storage costs associated with that retention. It allow sites to realize costs savings whilst retaining appropriate business access to data.
For more information about ILM refer to the ILM Planning Guide (Doc Id: 1682436.1) available from My Oracle Support.