Huawei unveiled its CloudEngine 16800 data center switch built for the Artificial Intelligence (AI) era.
The platform has three defining characteristics making it suitable for the AI era: an embedded AI chip, the capacity for a 48-port 400GE line cards per slot,, and the capability to evolve to the autonomous driving network.
Huawei said its embedded, high-performance AI chip will apply an innovative iLossless algorithm for the auto-sensing and auto-optimization of the traffic model. It promises lower latency and higher throughput based on zero packet loss. The company estimates its optimization will increase the AI computing power from 50 percent to 100 percent compared to traditional Ethernet, while improving the data storage Input/Output Operations Per Second (IOPS) by 30 percent. The CloudEngine 16800’s local intelligence and the centralized network analyzer FabricInsight creates a distributed AI O&M architecture capable of identifing faults -- a key goal of an autonomous driving network.
In terms of capacity, the Huawei CloudEngine 16800 data center switch boasts the industry’s highest density 48-port 400GE line card per slot, yielding an overall 768-port 400GE switching capacity. Huawei says power consumption per bit is reduced by 50% with the massive configuration. The company also claims to have overcome multiple technical challenges such as high-speed signal transmission, heat dissipation, and power supply.
http://e.huawei.com/topic/cloud-engine2019/en/index.html?ic_medium=hwdc&ic_source=ebg_banner_EEBGHQ179Q19W
