1. 黑斑羚：
Impala 是一个运行在 Hadoop 上的查询引擎。它是一个开源软件和大规模并行处理的 SQL 查询引擎。它支持内存数据处理。它率先使用 Parquet 文件格式，这是一种列式存储布局，针对数据仓库场景中典型的大规模查询进行了优化。它提供高性能、低延迟的 SQL 查询，还提供对以 Hadoop 文件格式存储的数据的交互式查询处理。
2. 蒙古包：
MongoDB 是一个跨平台的面向文档的非关系型(即 NoSQL)数据库程序。它是一个开源文档数据库，以键值对的形式存储数据。 MongoDB 由 MongoDB Inc. 开发，最初于 2009 年 2 月 11 日发布。它是用 C++、Go、JavaScript、 Python语言编写的。 MongoDB 提供高速、高可用性和高可扩展性。
Impala 和 MongoDB 的区别：
| S.No. | IMPALA | MONGODB |
|---|---|---|
| 1. | It is developed by Cloudera in 2013. | It is developed by MongoDB Inc. in 2009. |
| 2. | It is an open source software. | It is also an open source software. |
| 3. | Server operating systems for Impala is Linux. | Server operating systems for MongoDB are Solaris, Linux, OS X, Windows. |
| 4. | It do not support In-memory capabilities. | It support In-memory capabilities. |
| 5. | No transaction concepts. | ACID properties of transaction is used. |
| 6. | The replication method that Impala supports is Selectable Replication Factor. | The replication method that MongoDB supports is Master Slave Replication |
| 7. | All programming languages supporting JDBC/ODBC are supported by Impala. | It supports many programming languages like C, C#, Java, JavaScript, PHP, Lau, Python, R, Ruby, etc. |
| 8. | It support Sharding partitioning methods for storing different data on different nodes. | It support Sharding Partitioning methods. |
| 9. | JDBC and ODBC are used as APIs and access methods. | Proprietary protocol using JSON are used as APIs and other access methods. |
| 10. | The Primary database model is Relational DBMS. | The Primary database model is Document store. |