Andy Pavlo — Lessons learned from using machine learning to optimize database configurations
Ближайшая конференция — SmartData 2024, 4 сентября (online), 8–9 сентября, (Москва + трансляция).
Подробности и билеты: https://jrg.su/KFFBS7
— —
Database management systems (DBMS) expose dozens of configurable knobs that control their runtime behavior. Setting these knobs correctly for an application's workload can improve the performance and efficiency of the DBMS. But such tuning requires considerable efforts from experienced administrators, which is not scalable for large DBMS fleets. This problem has led to research on using machine learning (ML) to devise strategies to automatically optimize DBMS knobs for any application. Current research suggests that ML can generate better DBMS configurations more quickly than what is possible with human experts. And since these ML algorithms do not require humans to make decisions, they can also scale to support tuning thousands of databases at a time. Despite the advantages of ML-based approaches, there are still several problems that one must overcome to deploy an automated tuning service for DBMSs.
In this talk, Andy will discuss the challenges in using ML to optimize DBMS knobs and the solutions we developed to address them. His presentation will be in the context of the OtterTune database tuning service. Andy will also highlight the insights learned from real-world installations of OtterTune for MySQL, Postgres, and Oracle.
Видео Andy Pavlo — Lessons learned from using machine learning to optimize database configurations канала SmartData
Подробности и билеты: https://jrg.su/KFFBS7
— —
Database management systems (DBMS) expose dozens of configurable knobs that control their runtime behavior. Setting these knobs correctly for an application's workload can improve the performance and efficiency of the DBMS. But such tuning requires considerable efforts from experienced administrators, which is not scalable for large DBMS fleets. This problem has led to research on using machine learning (ML) to devise strategies to automatically optimize DBMS knobs for any application. Current research suggests that ML can generate better DBMS configurations more quickly than what is possible with human experts. And since these ML algorithms do not require humans to make decisions, they can also scale to support tuning thousands of databases at a time. Despite the advantages of ML-based approaches, there are still several problems that one must overcome to deploy an automated tuning service for DBMSs.
In this talk, Andy will discuss the challenges in using ML to optimize DBMS knobs and the solutions we developed to address them. His presentation will be in the context of the OtterTune database tuning service. Andy will also highlight the insights learned from real-world installations of OtterTune for MySQL, Postgres, and Oracle.
Видео Andy Pavlo — Lessons learned from using machine learning to optimize database configurations канала SmartData
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
Виктор Кесслер — Dremio SQL Lakehouse: Быстрые данные для всехНиколай Марков — Низкий уровень работы с даннымиJeff Zhang — Flink + Zeppelin: Streaming data analytics platformАлександр Крашенинников — Hadoop high availability: опыт BadooСаттар Гюльмамедов, Максим Мартынов — Эволюция ETL-инструментов на примере отдельно взятой Big DataДенис Ефаров — Hadoop 3: Erasure coding catastropheДмитрий Бугайченко — Stateful streaming: Кейсы, паттерны, реализацииБронислав Житников — NiFi-скрипты как элемент Less Code ETLSmartDataCoffeeКсения Томак, Максим Стаценко, Андрей Терехов — Разговор в телевизореВиталий Худобахшов — Имя — это фичаОльга Татаринова, Алексей Махоткин — Как моделирование данных повышает качество работы аналитиковВикторина «Наша игра»Павел Якунин — Безопасные интерактивные большие данные в банке: Business intelligence на ClickhouseАлексей Потапов — Глубокое обучение, вероятностное программирование и метавычисленияJacek Laskowski — Apache Spark as an in-memory-only data processing engine?Артур Хачуян — Построение кросс-ID для веб-аналитикиВладимир Озеров — Как устроено выполнение SQL-запросов в Presto/TrinoМихаил Марюфич — Надежные и масштабируемые пайплайны в OKЕвгений Ненахов (MTC Digital) — Организация потоковой обработки данных для Big Data