OLDR repositories are designed to handle older, historical data that no longer require modification. They offer built-in configuration retention and expiry rules as they are focused on delivering immutable data retention at limitless data volumes in combination with efficient online accessibility, all at the lowest possible price point. Accordingly, key attributes of OLDR repositories also include large-scale source data compression in order to save on physical storage, while still maintaining efficient retrieval performance across vast data sets.
Access to these data sets must also be supported in the context of cumulative data model schema changes over time. Since data retention should be unobtrusive to existing systems, standard access interfaces such as SQL and ODBC/JDBC are typically supported. Finally, OLDR repositories are designed to be particularly efficient for hosted environments or in public and private clouds, due to their scalability and ability to use resources elastically in a pay-for-use model, which allows for an optimal tradeoff between performance and cost.
The goal of an OLDR repository is to complement and support production applications by allowing OLTP and OLAP systems to 'go on a Big Data diet' by offloading static historical data without sacrificing on demand access. In this case, it makes sense to include OLDR as part of an integrated overall production transaction and data retention solution. In fact, leading ISVs who provide key applications for industries such as health care and financial services are actively reviewing OLDR repositories as a part of their Big Data retention offerings.
In other scenarios, where applications need to handle billions of machine-generated data records a day (e.g., Call Data Records, Events or Logs), extreme ingestion rates often dictate that OLDR repositories be used in place of traditional systems to store data that is immediately historical and should never be modified.
Like everyone, ISVs and MSPs are attracted to the compelling economics offered by OLDR repositories. Efficiencies include use of lower-cost commodity storage and servers, as well as a significant savings in personnel. OLDR systems are designed to run unattended and require little or no administration, allowing specialist resources to be redeployed. Total cost of ownership per terabyte when compared to traditional OLTP and OLAP repositories is significant enough that ISVs and MSPs have been able to create new revenue generating retention products and services at much higher margins. Downstream, their customers who rely on their turnkey applications benefit from an efficient and cost-effective way to meet their compliance obligations, thereby surfing the Big Data retention wave rather than being wiped out by it.