Rate and give feedback:
Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document.
Note: Even the most advanced machine translation cannot match the quality of professional translators.
Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
Scenario
Traditional networks store data on centralized storage servers, which become the bottleneck of system performance and are crucial to reliability and security, failing to meet the requirements of large-scale storage applications. In contrast, distributed networks use a scalable system structure to store data on multiple independent devices in load balancing mode. As such, the distributed network architecture is widely used on both front-end and backend networks. The distributed storage scenario has the following characteristics:
- A file is stored on multiple servers, which improves the file read/write speed and implements redundancy. The pressure on network performance is proportional to the storage performance requirements and affects the throughput of the distributed storage system. The intelligent lossless network needs to ensure high throughput in the distributed storage scenario.
- When applications write original and backup data to multiple servers, the 1:N incast traffic model is used. When an application reads a file and obtains data from multiple servers, the N:1 incast traffic model is used to aggregate data to the switch.
- Within the industry, most of the distributed storage systems use standard-byte traffic for communication. Therefore, there are large amounts of traffic of the same size on the network.