The Rivalry of Distributed Databases: A Deep Comparison of Doris and ClickHouse

# 1. Overview of Distributed Databases

A distributed database is a system that distributes data across multiple computers at different physical locations. This distributed architecture offers several advantages, including:

***Scalability:** Distributed databases can easily scale to handle increasing amounts of data by adding more nodes.
***High Availability:** If one node fails, other nodes can take over its workload, ensuring data is always available.
***Low Latency:** Distributed databases can place data close to the user's geographical location, reducing latency and improving performance.

# 2. Doris Database

### 2.1 Architecture and Principles of Doris Database

#### 2.1.1 Doris Database's Storage Model

Doris database utilizes a columnar storage model, storing data on disk by column. This storage model offers several advantages:

- **High Data Compression Rate:** Since columnar storage only stores data of the same type, it can use efficient compression algorithms to greatly increase data compression rates.
- **High Query Performance:** When queries involve specific columns, columnar storage can read only the relevant columns, avoiding unnecessary data reads, thus improving query performance.
- **Good Scalability:** Columnar storage can easily add or remove columns without reorganizing the entire dataset, making Doris highly scalable.

#### 2.1.2 Doris Database's Query Engine

Doris database uses Apache Impala as its query engine. Impala is an MPP (Massively Parallel Processing) query engine that can parallelize query tasks across multiple nodes, improving query performance.

Impala supports various query types, including:

- **Interactive Queries:** Supports low-latency interactive queries suitable for real-time analysis and data exploration.
- **Batch Queries:** Supports large-scale data processing tasks, such as ETL and data warehousing.
- **Real-Time Queries:** Supports real-time queries on streaming data, suitable for the Internet of Things and online analysis.

### 2.2 Advantages and Disadvantages of Doris Database

#### 2.2.1 Advantages of Doris Database

- **High Performance:** The columnar storage model and MPP query engine enable Doris database to have extremely high query performance.
- **High Compression Rate:** The columnar storage model can compress data effectively, saving storage space.
- **High Scalability:** Doris can easily scale up to hundreds of nodes to meet growing data volume and query demands.
- **Low Cost:** Compared to other commercial distributed databases, Doris database is open-source software, offering a cost advantage.

#### 2.2.2 Disadvantages of Doris Database

- **Lower Data Update Performance:** Due to the characteristics of the columnar storage model, Doris database has lower data update performance compared to row-based storage databases.
- **No Transaction Support:** Doris does not support transactions, limiting its use in certain application scenarios.
- **Weaker Data Consistency Guarantee:** Doris uses a eventual consistency model, which may lead to data inconsistency in certain cases.

# 3. ClickHouse Database

### 3.1 Architecture and Principles of ClickHouse Database

#### 3.1.1 ClickHouse Database's Storage Model

ClickHouse database uses a columnar storage model, storing data on disk by column. This storage model offers several advantages:

- **High Data Compression Rate:** Columnar storage can compress data of the same type, thus improving data compression rates.
- **Fast Query Speed:** Columnar storage avoids scanning unnecessary data during queries, thus improving query speed.
- **Good Scalability:** Columnar storage can easily scale to multiple nodes, enhancing the overall performance of the database.

The storage model of ClickHouse database mainly includes the following components:

- **Data Chunk:*