- Hands-On Big Data Modeling
- James Lee Tao Wei Suresh Kumar Mukhiya
- 180字
- 2021-06-10 18:58:53
Yet Another Resource Negotiator
Yet Another Resource Negotiator (YARN) is an example of an extension of the MapReduce framework. YARN is the architectural center of Hadoop and permits multiple data processing engines, such as interactive SQL, batch processing, and real-time streaming, to handle data stored in a single platform. It is known as the new generation of Hadoop.
YARN improves a Hadoop cluster in many ways. Some important features that are improved in YARN are listed as follows:
- Scalability: YARN has a ResourceManger, which has two components: a scheduler and an application manager. As the name suggests, the scheduler is responsible for allocating resources to the running application. The application manager is responsible for starting application masters and for monitoring and restarting them on different nodes, in the case of failures.
- Compatibility: YARN can run applications developed for Hadoop 1.x, without going through the modification process.
- Cluster utilization: YARN allocates clusters dynamically, over most static MapReduce rules.
- Multi-tenancy: YARN permits accessing engines, to use Hadoop as the common standard for each batch.
推薦閱讀
- 輕輕松松自動化測試
- 圖解PLC控制系統梯形圖和語句表
- PyTorch Deep Learning Hands-On
- JavaScript典型應用與最佳實踐
- 工業機器人運動仿真編程實踐:基于Android和OpenGL
- 基于敏捷開發的數據結構研究
- Linux系統下C程序開發詳解
- Mastering OpenStack(Second Edition)
- Hands-On Deep Learning with Go
- 案例解說Delphi典型控制應用
- 樂高創意機器人教程(中級 上冊 10~16歲) (青少年iCAN+創新創意實踐指導叢書)
- Kubernetes on AWS
- 基于Quartus Ⅱ的數字系統Verilog HDL設計實例詳解
- ARM嵌入式系統開發完全入門與主流實踐
- Flash 8中文版全程自學手冊