DDrona4U
Sign inCreate account
My learningBigQueryLesson 02
How does Dremel execution tree distribute a query across thousands of slots

Lesson 02

How does Dremel execution tree distribute a query across thousands of slots

Dremel runs queries through a multi-level tree of workers, where each level does partial work and passes results up. This is what lets a single SQL query fan out across thousands of slots in seconds.

Get the full lesson

Sign in to unlock everything beyond the preview — it's free.

  • Take timestamped notes as you watch
  • Read the full transcript and download resources
  • Join the discussion and track your progress
Sign inCreate free account

Curriculum

27 lessons · 2h 11m

0/27 lessons done2h 11m left
  1. 04:04

    How does bigquery separate storage and compute and why does this matter for sclaling

    04:04

  2. 06:51

    How does Dremel execution tree distribute a query across thousands of slots

    06:51

  3. 0307:08

    How does capacitor columnar format store data differently from row-based storage

    07:08

  4. 0403:53

    How does colossus distributed file system store and retrieve bigquery table data

    03:53

  5. 0506:35

    How does bigquery slot allocation work and how do on-demand slots differ from reserved slots

    06:35

  6. 03:17

    How does bigquery clustering physically rearrange data blocks to speed up filtered queries

    03:17

  7. 0703:48

    How does bigquery handle nested and repeated fields using STRUCT and ARRAY internally

    03:48

  8. 0806:16

    How does bigquery time travel work and how does it store 7 days of historical snapshots

    06:16

  9. 0904:49

    How does bigquery manage table snapshots and clones without duplicating storage

    04:49

  10. 1005:25

    How does bigquery external tables read data from cloud storage without importing

    05:25

  11. 05:43

    How does bigquery handle schema evolution when you add or remove columns from a table

    05:43

  12. 1203:28

    How does bigquery process a WHERE clause filter on a partitioned table with 10 billion rows

    03:28

  13. 1306:25

    How BigQuery Handles Data Skew During Shuffle Operations When One Key Contains 90% of the Data

    06:25

  14. 1406:41

    How BigQuery Query Cache Works and When It Serves Cached Results

    06:41

  15. 1503:10

    How BigQuery Uses Column Pruning to Skip Unused Columns and Reduce Bytes Scanned

    03:10

  16. 1600:19

    How BigQuery Uses Predicate Pushdown to Filter Data at the Storage Layer

    00:19

  17. 1701:27

    How BigQuery Materialized Views Automatically Rewrite Queries for Faster Execution

    01:27

  18. 1802:36

    How BigQuery BI Engine Uses In-Memory Caching to Accelerate Dashboard Queries

    02:36

  19. 1902:19

    How to Read and Optimize a BigQuery Execution Plan Using EXPLAIN and Query Statistics

    02:19

  20. 2004:20

    How BigQuery Processes a JOIN Between Two Large Tables Internally Step by Step

    04:20

  21. 2104:49

    How BigQuery Decides to Use Broadcast Join for a Small Dimension Table

    04:49

  22. 2205:06

    How BigQuery Handles Join Key Skew When One Customer Has Millions of Orders

    05:06

  23. 2304:04

    How BigQuery Optimizes a LEFT JOIN with NULL Handling Across Distributed Slots

    04:04

  24. 2409:30

    How BigQuery Executes RANK and ROW_NUMBER Over Billions of Rows Using PARTITION BY

    09:30

  25. 2504:58

    How BigQuery Processes LAG and LEAD Functions Across Distributed Partitions

    04:58

  26. 2607:04

    How BigQuery Calculates Running Totals Using SUM OVER with ORDER BY on Large Datasets

    07:04

  27. 2707:11

    How BigQuery Processes a QUALIFY Clause and How It Differs from HAVING

    07:11