ApsaraDB for ClickHouse: Distributed Real-Time Analytical Column Database Service

Quantitative Backtesting Analysis Based on Massive Historical Market Data

Financial companies engaged in quantitative trading typically use quantitative trading strategies and algorithms for investment. High-quality algorithms and high-quality quantitative backtesting are crucial for business profitability. Therefore, quantitative traders usually use the latest trading algorithms to test and review historical time-series data to determine profitability, a process known as backtesting. The complexity of the backtest model determines the number of parameters provided, such as initial capital, risk capital (%), portfolio size, commissions, average bid-ask spread, and benchmarks (e.g., S&P 500 index), helping traders evaluate the effectiveness of the latest trading strategies. Additionally, traders use current algorithms to backtest funds based on real-time trading conditions, track volatility, and predict future returns.

Solution Advantages

Efficient Real-Time Analysis Performance

Wide table query performance of ApsaraDB for ClickHouse is about 900 times that of MySQL and 10 times that of ElasticSearch.

Flexibility

MPP architecture of ApsaraDB for ClickHouse provides 50-200MB/s data writing capability per node, with linear scalability by adding nodes.

Rich Functions

ApsaraDB for ClickHouse offers 150+ built-in functions for high-precision calculations with time-series data using time window functions, facilitating time-series data aggregation analysis.

Low Cost

ApsaraDB for ClickHouse features cold-hot data tiered storage capability, allowing data archiving to low-cost storage, reducing storage costs by over 85%.

Customer Profile Analysis

Performs multi-dimensional analysis on user behaviors based on logs. Supports real-time business monitoring and real-time operations using screens

The Benefits of Our Solution

Efficient Analysis

Provides efficient analysis capabilities. Queries for 100 billion log entries are completed within 30 seconds. 95% of the queries for 100 million log entries are completed within 5 seconds.

Comprehensive Data Integration

Supports data imports from multiple big data services

Efficient Writing

Provides high throughput and supports writing 10 billion log entries per hour during peak periods

User Selection and Precision Marketing

Analyzes historical data and cleans real-time user behavior data. Allows you to select required users for marketing based on the historical tags and real-time tags to increase daily active users (DAU).

The Benefits of Our Solution

Efficient Analysis

Improves aggregation analysis performance 5 to 10 times over compared to traditional log analysis solutions. Queries for 100 million log entries are completed within milliseconds.

Efficient and Stable Writing

Prevents OOM problems when writing a large amount of log data compared with Elasticsearch. Provides a write speed of 50 to 200 MB/s

Advertising Evaluation and User Profiles

Performs multi-dimensional analysis on a large amount of user data, identifies users for advertising, and evaluates advertising costs. Creates user profiles, evaluates advertising accuracy, and estimates expected profits.

The Benefits of Our Solution

Efficient Analysis

Performs online analysis to identify users for advertising. Queries for 100 million data records in wide tables are completed within seconds.