With modern OLAP systems, you replace your Traditional OLAP Cubes one-to-one with another technology. Therefore, you keep everything the same on your current architecture but replace your cubes with a modern big data-optimized technology that focuses on the fastest query response time. See the Appendix for a comparison between modern OLAP Technologies.
Another term would be Real-Time OLAP systems, or Real-Time Data Warehouse.
Traditional open-source OLAP system, or Modern OLAP systems are:
# https://www.ssp.sh/brain/Modern%20OLAP%20Systems-20250127211718269.webhttps://www.ssp.sh/brain/Modern%20OLAP%20Systems-20250127211726579.webhttps://www.ssp.sh/brain/Modern%20OLAP%20Systems-20250127211738316.webpbphttps://www.ssp.sh/brain/Modern%20OLAP%20Systems-20250127211747214.webhttps://www.ssp.sh/brain/Modern%20OLAP%20Systems-20250127211652644.webhttps://www.ssp.shhttps://www.ssp.sh/brain/Modern%20OLAP%20Systems-20250127212018957.webpc="/brain/Modern%20OLAP%20Systems-20250127211652644.webp">
Source of above images.
To a certain extent, Semantic Layer and its Semantic Layer Tools are Cubes as well. Especially if they have an OLAP Cache Layer.
# DuckDB vs Modern OLAP system
DuckDB distinguishes itself on being lightweight, in-process, and thus highly portable. It is SQLite but for OLAP. Druid/ClickHouse/Pinot is a distributed system designed for large-scale data workloads. Rill use DuckDB for workloads that are 100GB or less. Michael Driscoll on LinkedIn: DuckDB is the world’s most downloaded analytics database, with 1+ million…
# Comparison
# Starrocks vs. ClickHouse/Druid
Ivan Torres says:
I use StarRocks open source on K8s and benchmarked it against Druid. I also used ClickHouse before. For real time data they all come pretty close, some optimizations of each database work better for specific use cases, but you can achieve pretty similar numbers with the three of them
What made me go to StarRocks is that it is a lot more versatile. It also does pretty good ad hoc analytics and integrates with external catalogs like Iceberg for reading external tables. You can also have full separation of storage and compute and achieve good performance with disk caching.
In Druid joining tables as hoc is pretty much impossible. You need a lot of work upstream for it to perform. In ClickHouse the integration to external tables is useless and you need to build everything by hand. Last, the commitment of StarRocks with open source is the best of all
Check out more on Traditional OLAP Cube Replacements and Traditional OLAP Cubes.
# Recommendations
Based on the fantastic write-up by Alireza Sadeghi, he suggests these choices:
Small-to-medium Deployments:
Overall ClickHouse is an excellent real-time OLAP engine suitable for small-to-medium environments. Its straightforward deployment, management, and architecture make it the preferred choice for general use cases.Large On-Premise Deployments:
For large-scale implementations, particularly on Hadoop or similar platforms, ClickHouse, Pinot, and Druid are leading candidates. The final selection should align with specific workload requirements and use cases.Cloud-Native Implementations:
Cloud-native deployments utilising object storage as the main persistence layer can leverage managed solutions like ClickHouse Cloud, or platforms such as StarRocks and Doris. However, consider that StarRocks and Doris introduced their decoupled architecture recently, suggesting careful evaluation for production readiness.Log Analytics & Time-series Data:
Druid and Pinot demonstrate particular strength in processing immutable time-series data, including web logs, machine logs, and clickstream events. Their support for hybrid tables makes them ideal for Lambda-style architectures.Unified Analytics with Query Federation:
ClickHouse, StarRocks, and Doris excel in unified analytics scenarios, offering query federation capabilities that enable seamless data access across diverse sources such as data lakes, lakehouses and DBMS systems.Hybrid Data Warehouse-OLAP Solutions:
StarRocks and Doris provide a middle ground, combining traditional data warehouse capabilities with real-time OLAP performance. They offer comprehensive CRUD operations, complex join support (including star schema), and ACID guarantees to some extend.
# Further Readings
Origin:
References: Vector Database (and Engine)
Created
