ClickHouse ● Incorporated in UK with distributed team in US/Canada/Europe ● US/Europe sponsor of ClickHouse community ● Offerings: ○ 24x7 support for ClickHouse deployments ○ Software (Kubernetes

ClickHouse ● Incorporated in UK with distributed team in US/Canada/Europe ● US/Europe sponsor of ClickHouse community ● Offerings: ○ 24x7 support for ClickHouse deployments ○ Software (Kubernetes

коротко Начните использовать ClickHouse сегодня! Вопросы? Можно сюда: › clickhouse-feedback@yandex-team.ru › Telegram: https://t.me/clickhouse_ru › GitHub: https://github.com/yandex/ClickHouse/ › Google

Sync Clickhouse with MySQL/MongoDB Company: Xiaoxin Tech. Industry: Education Team: Big Data Leader: wangchao@xiaoheiban.cn About 100 billion data this year till now 30 million users We use

Спасибо за внимание @kitaisreal Максим Кита Старший разработчик ClickHouse maksim-kita@yandex-team.ru

request #5590 and #7331 Begin Content Area = 16,30 21 UDF Development Explained Begin Content Area = 16,30 22 General Steps of UDF Development Design the Interface • Meta information Example: Will

we do ► Support JSONB DataType for tags & value ► Support LowCardinality(JSONB) ► Support BoolFilter skip index with JSONB data type ► Support TimeSeriesMergeTree Table Engine ► Support Multiple Streams Streams for AggregationFunction ► Support TimeSeriesAggregateFunction(store sum, min, max, avg) ► Support convert sum(time_series) to sum(time_series.sum) What we do QingCloud ChronusDB 青云 QingCloud

in ClickHouse › stochasticLinearRegression › stochasticLogisticRegression Stochastic methods do support multiple factors. That’s not the most important difference. 23 / 62 Stochastic linear regression CatBoost advantages › Good quality for default parameters › Sophisticated categorical features support › Models analysis tools 44 / 62 Gradient Boosting 45 / 62 Gradient Boosting 46 / 62 Gradient function) › Table function to sample data from table with repetition › Support for multiple features in simpleLinearRegression › Support for multiple loss functions in stochasticLinearRegression (MSE now)

in ClickHouse › stochasticLinearRegression › stochasticLogisticRegression Stochastic methods do support multiple factors. That’s not the most important difference. 23 / 62 Stochastic linear regression CatBoost advantages › Good quality for default parameters › Sophisticated categorical features support › Models analysis tools 44 / 62 Gradient Boosting 45 / 62 Gradient Boosting 46 / 62 Gradient function) › Table function to sample data from table with repetition › Support for multiple features in simpleLinearRegression › Support for multiple loss functions in stochasticLinearRegression (MSE now)

запросов с использованием GPU • Интеграция с ML & AI. Обработка графов • Batch jobs • Data Hub Support For Semistructured Data 27 JSO data type: CREATE TABLE games (data JSON) ENGINE = MergeTree; merge • Data is stored in columnar format: columns and subcolumns • Query nested data naturally Support For Semistructured Data 28 Example: NBA games dataset CREATE TABLE games (data String) ENGINE TABLE games (data JSON) ENGINE = MergeTree; SELECT data.teams.name[1] FROM games; — 0.015 sec. Support For Semistructured Data <-- inferred type DESCRIBE TABLE games SETTINGS describe_extend_object_types