Read the abstract ➤ https://www.conf42.com/DevOps_2024_Alexey_Ozeritskiy_distributed_sql_database_optimization Other sessions at this event ➤ https://www.conf42.com/devops2024 Join Discord ➤ https://discord.gg/DnyHgrC7jC Chapters 0:00 intro 0:38 preamble 0:51 about me 1:16 outline 1:51 yql: distributed sql database engine 3:35 massive data handling 4:28 yql architecture overview 5:48 example 6:08 execution plan 7:03 bechmark-driven approach 8:28 tpc-h benchmark 9:02 data generation 10:26 continuous integration (ci) 12:38 run distributed engine in one process 13:35 run distributed engine in multi-process configuration 15:15 unixbench's style measures 15:54 tpc--h 100: target values 17:10 hardware investigations 18:12 linus performance tools 20:00 more linux performance tools 21:58 slow join 25:04 perf top 25:42 stackcount 26:12 memory allocator 28:41 results containerization and performance 29:19 user code isolation 31:07 execution plan 31:47 executions pipeline with unix pipe 32:48 slow results 33:08 linux ipc performance 33:36 pipe performance 36:14 results: dq @pipe+ pipe 36:49 what's next? 37:49 thanks you