Kyvos on Google Cloud Platform
Applies to: Kyvos Enterprise Kyvos Cloud (SaaS on AWS) Kyvos AWS Marketplace
Kyvos Azure Marketplace Kyvos GCP Marketplace Kyvos Single Node Installation (Kyvos SNI)
This section explains how to deploy and configure Kyvos on Google Cloud Platform (GCP), including supported services, prerequisites, and setup steps to run Kyvos within a GCP environment.
Kyvos on Google Cloud Platform with No-Spark
Kyvos’ No-Spark architecture is deployed on Google Cloud Platform (GCP), which is designed for high-performance analytics with cloud-native scalability and without reliance on Apache Spark. The architecture integrates a variety of data sources, including Apache Iceberg, Parquet, CSV, BigQuery, Google Cloud Storage (GCS), and Snowflake.
At its core, the Elastic Processing Cluster (deployed via Google Cloud Instance Groups) and the Elastic Querying Cluster (running on the GC Engine) work together to process and serve queries on massive datasets efficiently. The Analytical Server, also running on the GC Engine, manages orchestration tasks such as metadata handling, job scheduling, and system coordination.
A Semantic Layer sits above the infrastructure to offer a business-friendly, performance-optimized view of the data. The Kyvos Data Storage layer stores both raw and aggregated data, which is leveraged by the Smart Aggregate Cache to accelerate frequently accessed queries. Users can interact with the system through industry-standard languages such as SQL, MDX, DAX, and LangChain, as well as REST/Java APIs. Business intelligence tools like Tableau, Strategy, Looker, Power BI, Excel, and SQL Server can seamlessly access Kyvos insights. Additionally, native tools such as Kyvos Dialogs, Kyvos Viz (powered by Gen AI), the Kyvos Excel Add-in, and Kyvos Reporting offer versatile interfaces for data interaction.
This GCP-based architecture is optimized for elasticity, concurrency, and intelligent caching, enabling organizations to gain faster, scalable, and AI-powered insights across large and complex datasets.
Kyvos on Google Cloud Platform with Spark
Kyvos enables self-service, interactive analytics on your Google Cloud Platform (GCP) by building a semantic layer directly on the cloud.
Kyvos consists of two main components: BI Servers and Query Engines. The Kyvos BI server and the query engines are deployed on Google Compute Engine.
Once the semantic data models are processed, they are stored in Google Cloud Storage for persistent storage. To achieve high performance, Kyvos replicates the cuboids and their metadata on the Google Persistent Disk. This helps deliver much higher performance than querying semantic models directly on Cloud Storage.
With Kyvos, you can also process semantic data models on your BigQuery data warehouse or directly on your Google Cloud storage.
Architecture
Kyvos' modern architecture enables deep integration in the GCP ecosystem. As you can see in the following figure, Kyvos also supports the Google BigQuery data warehouse on GCP.
Highlights
Elastic semantic model processing using Google Compute Engine
Leverages Google Cloud Storage for semantic model storage
High-performance, elastic querying
No need to move data out of your storage platform
Related topics