EditText
| BrowseFacets
| RecentChanges
| FindPage
|by John Sequeira
| Virtualization Wiki
|
http://www.hvr.biz
( was http://www.highvolumereplicator.com/ )
"HVR offers a robust solution for:
- Feeding a Hot-Standby Database. A Hot-Standby Database ensures a mission critical system remains available even after its main production databases crash. A 'switch-over' to the hot-standby database can be ad hoc or planned when maintenance work or upgrading would make the production database unavailable. This means TRUE 24x7 availability! HVR itself never needs system downtime, even when adding a new replicated target database.
- Feeding a Reporting Database. A Reporting Database provides an environment where long-running OLAP queries can be performed on real-time data without impacting performance on the production database. HVR not only replicates the regular tables of a database, it can also replicate the data directly into special unnormalized tables suitable for fast reporting (so-called 'reporting views'). This makes reports run even faster.
- Consolidation of Regional Databases is a task to which HVR is perfectly suited. Its speed and extremely efficient network compression techniques mean that multi-gigabyte databases can be replicated across continents and oceans. HVR’s wide range of functionality makes such feats possible. Collision handling can be important (a collision is when the same tuple is changed in two databases simultaneously). HVR's extended set of collision handling options allows five standard collision detection modes with four policies for collision resolution and four error handling severities. HVR's collision handling can exploit an application’s existing timestamps, which avoids the catastrophic performance effect of additional shadow tables for bi-directional replication.
- Bridging between Different Systems. HVR can replicate data between heterogeneous systems. These databases could have different DBMSs, because HVR replicates effortlessly between any combination of Ingres, Oracle or SQL/Server databases. The replicated databases do not need to have identical data models because HVR can perform data transformations during replication."
EditText
| BrowseFacets
| RecentChanges
| FindPage
|by John Sequeira
| Virtualization Wiki
|