CelerData enables enterprises to quickly and easily grow their business with the world’s most performant query engine for open lakehouses. Powered by StarRocks, CelerData delivers 3X the performance/cost of any other solution on the market and is the only platform uniquely designed to enable users to simplify their lakehouse architecture and ditch the need for a data warehouse. CelerData is used worldwide by market-leading brands including Airbnb, Pinterest, and Trip.com to generate critical new insights for these data-driven companies.
Rockset indexes looks stupid in traditional way from database perspective , but however in some way perfect for AI area. For the time they are changing!
🌟Join StarRocks on Slack: try.starrocks.com/join-starrocks-on-slack 🌟Airbnb Use Case: celerdata.com/blog/airbnb-builds-a-new-generation-of-fast-analytics-experience-with-starrocks
While Dremio explicitly forbids publishing performance tests or analysis against them, where you'll see the biggest fundamental difference is in the codebase. StarRock's is developed in C++ where Dremio is based on Java. Being C++ gives StarRocks the advantage of utilizing lower-level optimizations such as SIMD, which is crucial for fast OLAP queries. Additionally, StarRocks is fully open-source and actively maintained by a global community of contributors. In contrast, although Dremio is open-source, there has been no update for the past three months and currently, it does not support the submission of new issues.
StarRocks directly query data lake with its external catalog framework, no ingestion/data copying needed. Link to the docs: docs.starrocks.io/docs/sql-reference/sql-statements/data-definition/CREATE_EXTERNAL_CATALOG/
There is a huge performance difference. StarRocks' operators are fully vectorized so it is way faster for OLAP queries. Also query planning, StarRocks' cost-based optimizer can generate much more efficient query plans
@@SidaShen this is really promising, the reason why i asked is because we are using on premise cloudera solution, and impala is already configured, is there any official docs on setting up starrocks with cloudera
@@SidaShen great!, the reason i asked because we are running on premise cloudera solution, which impala is already configured, is there an official guide on setting up starrocks with cloudera?
@@abdullahmajed7554 Are you using Apache lceberg with Cloudera? If it is it should work. You can join the StarRocks channel and we can chat there about specific integrations bit.ly/starrocks-slack :D
Got error ERRcOR 1064 (HY000): Unexpected exception: Failed to create shards. error: INVALID_ARGUMENT:shard info can not be empty. when try create table, also starrocks-cn is not started.
@zamw4276 Good question, to be completely honest it'll really depend on your inifrastructure. Generally speaking, a good place to start would be this video by Tabular: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-G2YMCPdQfgM.htmlsi=BXMT0KRViUWuXzc7 but if you want a more personalized answer I'd suggest asking this question in the StarRocks Slack channel here: try.starrocks.com/join-starrocks-on-slack
Love starrocks! Do you guys have any plans to support Redpanda / Kafka as the storage layer? Especially for enterprises with the tiered aws storage and shadow indexing in Redpanda, I think Starrocks as the lakehouse engine would be absolutely amazing!
Thank you for your support and interest in StarRocks! We currently don’t have plans to integrate Redpanda/Kafka as the storage layer in our H1 or Q3 roadmap. If you haven't already, we invite you to join our StarRocks community on Slack - join.slack.com/t/starrocks/shared_invite/zt-2fou0ynxe-mZpcW54KewxWocehcBNoKQ to discuss this further and stay updated on future developments!
Useful Links: 🌟 Load Data into StarRocks from Any Database vis Sling: blog.slingdata.io/load-data-into-starrocks-from-any-database 🌟Join StarRocks on Slack: try.starrocks.com/join-starrocks-on-slack