StarRocks 3.3 is here, and it's more powerful than ever! In this video, we'll walk you through everything you need to know to get the most out of this release. Let's dive in and explore the new features and enhancements together!
-----------------------------------------------------------------------------------------------------------------------
00:00 Intro & Agenda
01:21 StarRocks Use Cases - Lakehouse Query Engine
03:07 StarRocks Use Cases - Real-Time Analytics Workloads
05:24 StarRocks 3.3: Shared-Data
05:36 Shared-Data: Fast Scheme Evolution
07:12 Shared-Data: Shared Data Manual Compaction
08:01 Shared-Data: Other
10:34 Q&A: Is there any downside to using Fast Schema Evolution? or is this strictly better than previous functionality?
11:11 Q&A: Is the Data Migration Tool a 1-time copy tool or does it sync data between clusters live continuously (CDC)?
11:34 Q&A: What is the query performance with Express One Zone? Is there an improvement?
12:19 Q&A: An implementation question, what protocol does Starrocks use for internode shuffling? Does it use grpc or something similar?
12:38 StarRocks 3.3: Data Lake Analytics
12:50 Data Lake Analytics: Data Cache-Cache Warmup
14:58 Data Lake Analytics: Data Cache-Others
16:41 Data Lake Analytics: Apache Iceberg Catalog - New Metadata Framework
19:17 Data Lake Analytics: Apache Paimon Catalog, Clcikhouse Catalog and Kudu Catalog
22:09 Q&A: Any support for Avro files ingestion yet?
22:31 StarRocks 3.3: Materialized View
22:34 Challenge with Pre-Computation Pipelines
24:20 StarRocks Materialized View
25:45 Materialized View Rewrite - View-Based Rewrite
27:26 Materialized View Rewrite - Aggregated Push-Down
28:38 Materialized View Consistency - When Querying MV Directly
30:01 Materialized View Consistency - Data Freshness vs. Performance
31:05 Partitioned Materialized View -Enhancements
32:55 Optimizer - Materialized View Performance
36:06 Q&A: Assuming that the materialized view supports multiple tables, if at 7:00 PM the materialized view builds 100 rows based on the state of the tables, and at 7:01 PM the underlying tables change, resulting in 10 records being updated in the materialized view, as a consumer of the materialized view, will I have timestamp support that shows 7:01 PM for those 10 rows?
37:46 StarRocks 3.3: Query and Storage
37:48 Query and Storage- Data Processing: Spill to Disk GA; Temporary Table
39:37 Query and Storage-Data Processing: Grouped Execution for Collocated Groups
40:47 Semi-Structured Data- JSON
41:53 Storage - Expression Partition
42:41 Storage - Indexes: Ngram Bloom Filter; Inverted Index;
44:03 AWS Graviton Support
44:56 StarRocks Community Updates
46:33 Q&A: Is the enable_spill set in fe.conf?
46:47 Q&A: Do you guys have any webinars planned for teaching about optimizing partitioning? See StarRocks Best Practices: Data Modeling - www.starrocks.io/blog/starroc...
-----------------------------------------------------------------------------------------------------------------------
Learn more at starrocks.com/
Connect with us:
LinkedIn: / celerdata
Twitter: / celerdata
CelerData Website: celerdata.com/
StarRocks GitHub: github.com/StarRocks/StarRocks
StarRocks Website: www.starrocks.io/
Slack: try.starrocks.com/join-starro...
#DataAnalytics #DataEngineering #DataLakeAnalytics #OLAP #DataAnalyst #DataEngineer #DataInfrastructure #UserFacingAnalytics #Database #AnalyticalDatabase #DataLake #DataLakeHouse #DataWarehouse #DataScience
Негізгі бет Ғылым және технология StarRocks 3.3 is Here: Key Features and Improvements
Пікірлер: 2