This feature is implemented in 1C:Enterprise version 184.108.40.2068.
We continue to develop the mechanism for export/import of configurations to/from XML files. In one of the previous versions of 1C:Enterprise, we implemented partial import of configurations from XML files. Now we have completed the reverse operation – partial (incremental) export of configurations.
As a result of these two changes, group development of large configurations should become easier and faster.
This feature is implemented in 1C:Enterprise version 220.127.116.118.
In version 8.3.9, we implemented a significant number of tasks to optimize different mechanisms of the platform. We would like to tell you about one of them in this article. The task in question concerns improved performance of web services.
Reuse of sessions
Insufficient performance of web services was due to the fact that each web service call had significant "overhead expenses" related to creating and terminating a session. Besides, every time a session was created, the handler SessionParametersSettings() was executed, which could be quite "heavy" in a standard configuration.
In addition, there was a functional deficiency. Web services did not have a state. This prevented us from implementing the logic that uses saved state between web service calls.
In version 8.3.9, we fine-tuned the mechanism for web services (SOAP services, HTTP services, and OData services). As a result, their performance has improved by approximately 10 times.
Deploying 100-users 1C:DocFlow solution in Vietnam International Bank (VIB) we had to choose servers hardware characteristics for the system. We used the official 1C recommendations that can be find in this tech article. The main idea behind the article is that we can extrapolate the hardware component utilization of some live system (called Model) to our system, assuming that the model system users utilize the hardware components to the same extent as our users will.
One of the most important issues solved for Vietnam International Bank (VIB) was severe performance insufficiency reported by everyone who tried to use the system during its acceptance testing. Literally every user action (like opening a form or refreshing a list) used to take a few seconds while the more massive operations (like posting a heavy document) could take up to 30 seconds.