Release Notes - What's new in Kyvos 2024.12
Here is the list of features and updates included in the Kyvos 2024.12 release.
Features and Enhancements for 2024.12
Kyvos 2024.12 release brings the following new features, enhancements, and bug fixes.
Conversational Analytics Using Gen AI-Powered Kyvos Dialogues Enhancements
This release comes with the following Conversational Analytics enhancements:
Support for exporting a card from a chat session in Image or CSV format.
Support for executing commands, such as questions about the semantic model or other topics, excluding business-related questions.
Support for applying the data formatting (units) for SQL calculated fields based on the context of the question asked in the chat.
Support for exporting pinned chats and full pinboard in PDF format.
Optimize the Retrieval-Augmented Generation (RAG) process to minimize the number of columns sent to the language model (LLM) for query generation.
Kyvos Modern Architecture (No-Spark Ingestion) Enhancements
This release comes with the following no-Spark enhancements:
Support for creating levels of date data type from the ‘Date’ column.
Support for formatting on levels and attributes with underlying field type as numeric. This is applicable only for Kyvos Viz.
Support for decimal data type for level and attribute in dimensions.
Support for filtering calculated members will enable users to select one or more calculated members in the filter list, just like other data members by setting the value of the kyvos.query.calculated.member.filter.enabled property to true. This capability allows users to filter and customize views based on their needs while enhancing Kyvos' complex data analytics.
Refined User Interface (UI) for the processing of metadata for attributes. This enhancement improves clarity and usability, enabling users to understand the metadata process easily.
Support for processing Parent-Child hierarchy.
Enhancements in the single table semantic model:
Support of calculated measure in flexible query
Support of distinct count measure in MDX on raw data value
Support of calculated members with flexible queries
Kyvos SaaS Enhancements
This release comes with the following Kyvos SaaS enhancements:
Support for deleting invited users within the account now exists. This applies to Portal users and Admins, but they cannot delete the owner of the account, ensuring that the highest level of control remains secure.
Support for downloading payment gateway transaction receipts from the ‘Transaction History’ tab.
Support for dynamic time zone. This feature automatically detects the user's time zone based on their device settings or IP address, ensuring that all time-related data is relevant to their local time. Once the user's time zone is identified, the application converts all time values accordingly. This includes timestamps for cluster creation, audit logs, transactions, and any other time-sensitive information.
Support for Oracle Analytics Cloud (OAC) connection
Kyvos now supports Oracle Analytics Cloud (OAC) connection for querying semantic model.
Support for ROLAP queries on Postgres data source with Spark
Users can now query the semantic model designed over the Postgres source without processing it using the old architecture, i.e., with Spark.
Kyvos Manager Enhancements
This release comes with the following Kyvos Manager enhancement:
Thread Dump Management: Kyvos now supports generating and downloading thread dumps directly from the Logs Files tab, available on the Monitor page. With this release, users can:
For Kyvos Services (Query Engine and BI Server): Download, view, or delete thread dumps from the Services Logs section.
Kyvos Manager Services: Generate, download, view, or delete a thread dump for the Kyvos Manager service from the Kyvos Manager tab.
NOTE: The support for generating a thread dump for KM Agent has already been supported in earlier release.
Cloud Support Enhancements
This release comes with the following cloud enhancements (AWS, Azure, and GCP):
Configure Kubernetes node pool as dedicated or shared: Kyvos now supports configuring Kubernetes node pools as either dedicated or shared, which determines the capabilities available on that node pool.
Version-Based Patch Deployments: In cloud-based environments, users can now perform version-based patch deployment either exclusively for Kyvos Manager or for both Kyvos Manager and Kyvos, using a single-click patch deployment.
AWS Enhancements
Eliminate the Need for Bundle Access Credentials: With this release, for AWS Managed Services and SaaS-based deployments, patches and upgrades no longer require S3 bucket access credentials to download Kyvos bundle.
Amazon Linux 2023 (AL2023) Operating System: Kyvos now supports the Amazon Linux 2023 operating system. This operating system is applicable for AWS Managed Service and Enterprise.
Azure Enhancements
Azure (ABFS) Reading for Quering on Kyvos Analytical Store: With this release, Kyvos is certified for ABFS reading for querying on Azure using the Kyvos Analytical Store.
Automated Cluster Deployment with Pre-Configured System-Assigned Managed Identity: Kyvos now supports automated cluster deployment in environments where a system-assigned managed identity is pre-configured on Virtual Machines prior to Kyvos deployment.
GCP Enhancements
Google File System (GFS) Reading for Quering on Kyvos Analytical Store: With this release, Kyvos is certified for GFS reading for querying on GCP using the Kyvos Analytical Store.
Kyvos Viz. Enhancements
This release comes with the following Kyvos Visualization enhancements:
Separate Email Notification for Failed and Canceled Semantic Models: Kyvos now provides separate email notifications for failed and canceled semantic models, enabling users to clearly differentiate between these notifications with ease and clarity.
Rendering measures on X-Axis of XY charts: Users can now plot measures on the X-Axis for XY charts by enabling the Plot Measures on X-Axis property.
Usability Enhancements
This release comes with the following Kyvos usability enhancements:
Dataset Preview Enhancements: Kyvos now offers users the flexibility to preview the entire dataset, replacing the previous limitation of a partial dataset preview.
NOTE: Opting to preview the full dataset may result in higher-than-expected costs. Additionally, the execution time required to generate the dataset preview could increase, depending on the size of the data.Enhanced Materialization Settings for Attributes: Users can now experience improved materialization settings for attributes in semantic model.
Format Numeric Values with Units: Users can now format numeric fields by applying display units such as thousands (K), millions (M), billions (B), or trillions (T). Numeric formatting can be applied at both the worksheet level and the semantic model.
Recurring Job schedule for Multiple Days: Users can now configure semantic model job schedules for specific days of the week using a recurring schedule. This includes flexible options to schedule jobs on non-consecutive days, such as alternating days, specific weekdays, or any custom combination of days.
API Enhancements
This release comes with the following API enhancements:
New API Enhancements for Kyvos Manager
The following new API has been added.
Status of the given Service: To get the status of the given service.
Status of the all services on given node: To get the status of all services on the given node.
Existing API Enhancements
New parameters have been added to the following existing REST API:
API name | New Parameter Name | Parameter Description |
---|---|---|
get Job status | getBuildNumber | If the value of the ‘getBuildNumber’ parameter is 'true', the process number of the semantic model process is also provided in the response. |
For further details, please see the REST API guides.
Deprecated Property
With this release, the WEB_DEPLOYMENT_TYPE in olapengine.properties has been deprecated.
Note
This document covers generic release notes for all Kyvos editions. Depending upon your deployment type, some of these may not be relevant to your environment.
Read more
Copyright Kyvos, Inc. All rights reserved.