High performance |
||
From DataObjects.Net Wiki
|
(Redirected from Peak performance)
DataObjects.NET designed to operate with large volumes of data under high load. It is the only ORM allowing the application to „crunch“ unlimited amounts of data even in a single transaction. Batched key generation (a version of Hi\Lo algorithm), automatic command batching, compiled and future (delayed) queries, prefetch API supporting arbitrary prefetch paths, „smart“ collections of entities (smart EntitySets), entity type caching (increasing data access performance for polymorphic hierarchies) and boolean branching are just some of features allowing DataObjects.Net to be noticeably faster than most of competitors in typical scenarios. That’s why we don’t fear direct and open comparisons: Xtensive is the only ORM vendor that have implemented a test suite measuring quality of LINQ implementation and performance of basic data access operations for 8 different ORM products: ORMeter.net, which is now supported by commercial ORM product vendors scoring their product there. It should be noted that now we aren’t leaders in ORMeter.net performance tests. Pretty complex DataObjects.Net currently has no weapon against the simplicity of the current favorite of this test. Since ORMeter.net tests are focused on measurement of peak performance on simple data access operations, many of benefits sophisticated ORM products offer simply can’t play their role there, so this was predictable. But we have no doubt that within the next 3 months we will be able to „push“ the current favorite there, despite its simplicity. In the meantime, we are pleased with best result in LINQ quality test and definitely good result in performance test. See an article High performance in DataObjects.Net Manual. |



