🥇Release InterSystems IRIS 2019.1

In mid-March was released the new version of the InterSystems IRIS data platform 2019.1

We present to your attention the list of changes in English. A complete list of changes and Upgrade Checklist in English can be found at this link.

Improvements in InterSystems Cloud Manager

InterSystems Cloud Manager is a utility for easily deploying InterSystems IRIS installations in the cloud. The 2019.1 release includes the following features:

Support for availability zones. Creation of installations that occupy multiple zones within a single region. More details — "Deploying Across Multiple Zones».
Support for asynchronous mirroring. Inclusion of asynchronous mirrored nodes in the installation configuration. More details — "Mirrored Configuration Requirements».
Installation of InterSystems IRIS without using containers, directly from the installation package. More details — "Containerless Deployment».
Support for service discovery. More details — Sharing ICM Deployments.

Client languages

The release includes new modules for working with InterSystems IRIS:

IRIS Native API for .NET — low-level access to globals from .NET applications. More details — "First Look: InterSystems IRIS Native API for .NET».
Relational access from Python — support for using pyodbc. More details — "Support for pyodbc Python ODBC bridge».
Dynamic Java Gateway — calling Java programs from InterSystems IRIS without creating proxy classes. More details — "Using Dynamic Object Gateways».
Using shared memory when working with Java Gateway.
Compatibility with Hibernate 5.2, 5.3. More details — "Hibernate Support».
A new Java utility for bulk data loading. More details — "The Simple Data Transfer Utility».

Improved scalability and management of distributed cluster

The distributed cluster in InterSystems IRIS shares data and cache among multiple servers, providing flexible, low-cost scalability for queries and data additions. This release includes the following improvements:

Support for more scenarios for SQL. Nodes in the cluster can now be added at any time, regardless of the database schema and used keys. After adding a node, data can be re-sharded (offline). More details — "Rebalance Sharded Data Across Additional Shard Data Servers».
A new overview and configuration page for the cluster has appeared in the Management Portal.
A new API for creating coordinated backups of the cluster. More details — "Coordinated Backup and Restore of Sharded Clusters».
A new Java utility for bulk data loading has also been optimized for cluster operations.

Improvements in SQL

This release includes significant enhancements in SQL performance and usability.

Auto-parallelization of suitable queries. Read more — «System-Wide Parallel Query Processing».
A new TUNE TABLE command for tuning a table via the SQL interface. Read more — «TUNE TABLE».
Improvements in SQL Shell, which now allows viewing schemas, tables, and views defined or accessible in the current namespace. Read more — «Using the SQL Shell Interface».
Viewing the query plan now shows subplans of composite plans during parallel processing and cluster queries.
You can now add options to the query text to override system SQL settings for this query. Read more — «Comment Options».
In each release, InterSystems includes various subtle SQL improvements that are not apparent to applications. In 2019.1, especially many such improvements were added to the query optimizer and code generator. Together with automatic query parallelization, this should significantly enhance the performance of applications using InterSystems IRIS SQL.

Improvements in Analytics

The ability to set partial dates in Business Intelligence. For example, specifying a date for which only the year or the year and month are known. Read more — «Partial Dates».
A new %SQLRESTRICT construct for filtering data through SQL within an MDX query.

Improvements in integration capabilities

This release contains many enhancements that simplify configuration and troubleshooting in productions:

Searching and viewing all pathways a message may take in production. Read more — «Viewing Interface Maps».
Finding locations where production components reference other production components. Read more — «Finding Interface References».
Testing data transformations. In the testing dialog, you can now set values for aux, context, and process objects as if the transformation was called with initialized objects. Read more — «Using the Transformation Testing Page».
DTL Editor. New actions — switch/case. The ability to group actions and add comments to transformations.
You can now send a message to a rule and see the execution result without running the message through the entire production. Read more — «Testing Routing Rules».
The ability to download messages from the Message View to a local computer. Learn more — «Exporting Messages».
The ability to download log events to a local computer. Learn more — «Introduction to the Event Log Page».
The rules editor now allows comments to be added to the rules and enables opening and editing transformations that are used in the editable rule.
The configuration for the 'Queue Wait Alert' now specifies the time after which a notification is generated for a message that is either in a production item queue or an active message. Previously, this timeout only applied to messages in the production item queue. Learn more — «Queue Wait Alert».
Restriction of access to 'Default System Settings'. Administrators can configure users for editing, viewing, or deleting default settings. Learn more — «Security for System Default Settings».
The ability to export production items to a local computer. Learn more — «Exporting a Production».
Deployment of production from a local computer is now possible. Learn more — «Deploying a Production on a Target System».
Enhanced navigation on the production settings page. Links have been added to the tabs on the production settings page for quick opening of corresponding items in a separate window. On the Queue tab, clicking on the message number opens the trace. On the Messages tab, clicking on the session number opens the trace. On the Processes tab, clicking on the message number opens the trace, while clicking on the process number opens the process details window.
New options in the business production item addition wizard. Users can now automatically assign default system values if fields are left blank and set a package prefix for generating routing rules. Learn more — «Wizard Options».

System performance and capabilities

Significant improvements in scalability and performance, particularly for large systems with non-uniform memory access (NUMA). These enhancements include scalability changes in statistics collection and global buffer management, performance improvements in global mapping at the index level (subscript-level mapping), and other optimizations to avoid pointer-block traversal. To enable these improvements, modifications were made in the system statistics and memory usage, as described in the checklist for this release. These improvements increase the memory size allocated for global buffer metadata by 64 bytes per buffer on Intel systems and by 128 bytes on IBM Power. For example, for an 8-kilobyte block buffer, the growth will be 0.75% for Intel systems. These improvements also led to minor changes in the display of statistics within utilities and the Management Portal.
Key Management Interoperability Protocol (KMIP). Starting with this release, InterSystems IRIS can function as a client for an industrial key management server. KMIP, an OASIS standard, empowers centralized key management. You can use keys from the KMIP server for encrypting both databases and individual elements. Keys from the KMIP server are accessible just like keys stored in files, for example, for encrypting log files. InterSystems IRIS supports copying keys from the KMIP server to local files, allowing for local backups. More details are available inManaging Keys with the Key Management Interoperability Protocol (KMIP)»
A new DataMove utility for transferring data from one database to another while simultaneously changing global mapping settings. More details inUsing DataMove with InterSystems IRIS».
Support for strings longer than 3,641,144 in JSON objects.
Support for connecting IRIS Studio with Caché and Ensemble.
Support for SPNEGO (Microsoft Integrated Windows Authentication) protocol during HTTP connections. Now %Net.HttpRequest can utilize Windows authentication via HTTP 1.1 to connect to a secured server. Users provide access credentials, or %Net.HttpRequest will attempt to use the current context. Supported authentication schemes are Negotiate (Kerberos & NTLM), NTLM, and Basic. More details are available inProviding Authentication».
Improvement of logging performance and asynchronous input/output.