Optimization




Functionality After Before Result of changes
Development tools.
The performance of the following Designer operations is improved:
  • Module check
  • Configuration check
The performance of module and configuration checks was insufficient. Improved usability for developers of large applied solutions.
1C:Enterprise script.
The performance of 1C:Enterprise script execution is slightly improved in the thin and thick client applications. The performance of 1C:Enterprise script execution in client applications was insufficient. Improved performance of 1C:Enterprise script execution in the thin and thick client applications.
Common objects.
The performance of updating value table column cells in columns that are not a part of any value table index is improved. Redundant rebuilding of value table indexes was performed in the event of updating a cell in a column that was not a part of any value table index. Improved performance for bulk value changes in columns that are not part of any value table index.
Data exchange.
Server clusters.
The reading of accumulation register, information register, and calculation register record sets during the creation of an initial image of a client/server infobase is optimized.

The time required to create an initial image of a client/server infobase is significantly reduced.

The performance of creation of an initial image of a client/server infobase was insufficient. Improved performance of infobase preparation for new distributed infobase nodes.
Command interface.
Development tools.
Improved performance of opening the configuration command interface editor form. The performance of opening the configuration command interface editor form was insufficient. Better usability of editing the configuration command inerface.
DBMS operations.
Server clusters.
The algorithm of 1C:Enteprise server cluster operations with PostgreSQL DBMS is modified. This reduces PostgreSQL memory usage.

If a 1C:Enterprise server cluster connection to PostgreSQL DBMS is idle for 30 minutes, it is released.

The algorithm of 1C:Enteprise server cluster operations with PostgreSQL DBMS caused increased PostgreSQL memory usage.

1C:Enterprise server cluster did not release idle connections to PostgreSQL DBMS.

Reduced PostgreSQL memory usage for operations with 1C:Enterprise server cluster.
Development tools.
Infobase configuration update is optimized. Configuration comparison in scenarios where objects are mapped by their names is optimized. The performance of infobase configuration update was insufficient. Improved performance of configuration updates.
DBMS operations.
Temporary table operations are optimized in scenarios that use PostgreSQL DBMS. The performance of temporary table operations in scenarios that used PostgreSQL DBMS was insufficient. Improved performance in multiuser mode in scenarios that use PostgreSQL.
Web services and HTTP services.
Sending large files (tens and hundreds of megabytes) over HTTP services is optimized. The performance of sending large files (tens and hundreds of megabytes) over HTTP services was insufficient. Improved performance of sending large files over HTTP services.
DBMS operations.
Optimized temporary table operations in scenarios that use Microsoft SQL Server and include high load on 1C:Enterprise and active usage of temporary tables. Pauses during the closing of DBMS connections are eliminated. The performance of temporary table operations in scenarios that used Microsoft SQL Server and included high load on 1C:Enterprise and active usage of temporary tables was insufficient. Improved performance in multiuser mode in scenarios that use Microsoft SQL Server.

DBMS operations.
In the file mode, queries are optimized for scenarios where an applied solution has separators but actually contains a single data area.

To apply this optimization to an existing infobase, perform infobase verification and repair with the Reindex infobase tables check box selected. Once the restructuring is completed, the optimization is applied to all restructured objects.

In the file mode, in scenarios where an applied solution had separators but actually contained a single data area, the performance of queries was insufficient. Improved file infobase performance in scenarios where an applied solution has separators but actually contains a single data area.
Development tools.
Command interface.
Reduced time of displaying changes in configuration objects that affect the configuration command interface. Such changes cause the immediate update of the active form editor, while other form editors are updated when they are activated. Changes in configuration objects that affected the configuration command interface caused the update of all open form editors. Improved Designer performance during the update of properties of configuration objects that affect the command interface.
Formatted documents.
The performance of formatted documents is improved. The performance of formatted documents was insufficient. Improved performance of operations with large texts.
Managed locks.
Absorption of managed locks is implemented (when two managed locks are applied at the same time, only one will take effect). A lock that does not have all lock areas specified absorbs a lock that has more lock areas specified, provided that the values in their common areas match.

If lock escalation causes a conflict with other locks, the escalation is not performed and an attempt to set the requested lock is made instead. This can lead to a total of more than 100 000 locks in a single area.

In scenarios where a single transaction uses a large number of managed locks with slight differences (for example, a single dimension varies), it is recommended that you set a single managed lock where the value of that dimension is not set.

Lock absorption was not performed. If lock escalation caused a conflict with other locks, the transaction that attempted to escalate the lock was canceled. Reduced number of locks and reduced amount of memory used by the cluster manager.

Improved system reliability in scenarios with large numbers of simultaneous locks.

DBMS operations.
Calculation registers.
When deleting records belonging to the actual action period, recording of register sets of calculation registers that support actual action period is performed within several queries, which reduces the chance of escalating the lock to the table level. This does not increase the record set execution time. When deleting records belonging to the actual action period, recording of register sets of calculation registers that supported actual action period was performed within a single query, which could cause lock escalation to the table level. Reduced chance of table lock, which gives more options for performing parallel operations.
Calculation registers.
The index of the actual action period table of a calculation register is complemented by the start date and end date fields, and the index is now clustered. This reduces the chance of lock escalation to the table level when writing record sets of calculation registers that support action period.

For the chart of calculation types table that stores serial numbers of calculation type displacements, a nonunique index by displacement number is implemented. As a result, simultaneous editing of a single chart of calculation types in different data areas of a separated infobase never requires waiting for lock releases.

To apply this optimization to existing infobases, perform the infobase verification and repair operation with the Restructure infobase tables option selected.

The index of the actual action period table of a calculation register did not include the start date and end date fields, and the index was not clustered.

For the chart of calculation types table that stored serial numbers of calculation type displacements, a nonunique index by displacement number was not available.

Improved calculation register performance in large and separated databases.

In large databases and especially separated databases, reduced number of table-level locks for the following tables: actual action period table of a calculation register and table that stores serial numbers of calculation type displacements.

Managed forms.
Development tools.
The first opening of managed form editor in Designer is faster. The first opening of managed form editor in Designer took significant time. Reduced waiting time in managed form development.
Information registers.
Faster writing of information register record sets that have slice first or slice last options enabled.

Faster execution of SliceFirst() and SliceLast() information register manager methods.

The performance of writing information register record sets that had slice first or slice last options enabled and the performance of SliceFirst() and SliceLast() information register manager methods was insufficient. Improved performance of information registers with slice first or slice last options enabled.
Interface.
Managed forms.
Functional options.
The performance of RefreshInterface(), SetInterfaceFunctionalOptionParameters(), and SetFormFunctionalOptionParameters() methods is improved. Delays introduced by these methods in applications with multiple simultaneous user connections are eliminated. The performance of RefreshInterface(), SetInterfaceFunctionalOptionParameters(), and SetFormFunctionalOptionParameters() methods was insufficient. The performance was reduced in applications that called these methods in scenarios with multiple simultaneous user connections. Streamlined operations in applications that use functional options.
Managed forms.
Mobile applications.
In the mobile platform, the performance of operations with large spreadsheet documents is improved. In the mobile platform, the performance of operations with large spreadsheet documents was insufficient. Improved performance of operations with large spreadsheet documents on mobile devices.

Comments
0
Add comment