Overview<\/a> | Features<\/a> | Compatible<\/a> | Downloads<\/a> | Solutions<\/a> | Resource<\/a> | Back<\/a><\/p>\n<\/div><\/div> AI \/ ML | AI Switch<\/a><\/p>\n<\/div> | AIS800-64D<\/p>\n<\/div><\/div><\/div><\/div><\/div> 51.2T DATA CENTER SWITCH<\/p>\n<\/div> 64 x QSFP-DD800 switch ports with Tomahawk 5 Overview<\/p>\n<\/div> The Edgecore AIS800-64D is a high-performance, low-latency switch for high-performance data centers.<\/p>\n Application Scenarios:<\/p>\n<\/div> Standards-based (Ethernet) networking for AI\/ML training, leveraging low latency and high-throughput RoCEv2. Reduces Job Completion Time (JCT) using the cognitive routing and congestion management capabilities of the switch. Fully programmable telemetry enables sophisticated on-chip applications for heightened network insight and efficient network management.<\/p>\n<\/div><\/div><\/div> The large number of high-capacity Ethernet ports enables server interfaces to transition to higher speeds and denser networks. Enables the virtualization of compute and storage with VxLAN switching and routing.<\/p>\n<\/div><\/div><\/div> Support 400G QSFP-DD ZR\/OpenZR+\/+6dBm ZR+ and future proof 800G OSFP ZR\/ZR+\/+6dBm for Cloud DCI scenario.<\/p>\n<\/div><\/div><\/div>AIS800-64D<\/h1><\/div>
\nHigh-performance, low-latency switch for high-performance data centers<\/p>\n<\/div><\/div><\/div>![](\"https:\/\/www.edge-core.com\/wp-content\/uploads\/2024\/12\/AIS800-64D-A.png\")
<\/span><\/a><\/li>![](\"https:\/\/www.edge-core.com\/wp-content\/uploads\/2024\/12\/AIS800-64D-F.png\")
<\/span><\/a><\/li>![](\"https:\/\/www.edge-core.com\/wp-content\/uploads\/2024\/12\/AIS800-64D-FF.png\")
<\/span><\/a><\/li>![](\"https:\/\/www.edge-core.com\/wp-content\/uploads\/2024\/12\/AIS800-64D-B.png\")
<\/span><\/a><\/li>![](\"https:\/\/www.edge-core.com\/wp-content\/uploads\/2024\/12\/AIS800-64D-BB.png\")
<\/span><\/a><\/li><\/ul><\/div><\/div><\/div><\/div><\/div><\/div>![\"Audi\"](\"https:\/\/www.edge-core.com\/wp-content\/uploads\/2023\/10\/77-AI-ML-100x100-1.png\" "\\"77-AI-ML-100×100\\"")
AI\/ML use case<\/h6><\/div><\/div><\/div><\/div><\/div>
![\"Audi\"](\"https:\/\/www.edge-core.com\/wp-content\/uploads\/2023\/10\/75-Low-Latency-100x100-1.png\" "\\"75-Low-Latency-100×100\\"")
Low Latency<\/h6><\/div><\/div><\/div><\/div><\/div>
![\"Audi\"](\"https:\/\/www.edge-core.com\/wp-content\/uploads\/2024\/07\/81_30W-Max-100x100-1.png\" "\\"81_30W-Max-100×100\\"")
Transceiver (max 30W) <\/h6><\/div><\/div><\/div><\/div><\/div>
![\"Audi\"](\"https:\/\/www.edge-core.com\/wp-content\/uploads\/2023\/10\/79-High-Radix-100x100-1.png\" "\\"79-High-Radix-100×100\\"")
High Radix<\/h6><\/div><\/div><\/div><\/div><\/div>
![\"Audi\"](\"https:\/\/www.edge-core.com\/wp-content\/uploads\/2023\/10\/78-End-to-end-congestion-control-100x100-1.png\" "\\"78-End-to-end-congestion-control-100×100\\"")
End-to-end congestion control<\/h6><\/div><\/div><\/div><\/div><\/div>
![\"Audi\"](\"https:\/\/www.edge-core.com\/wp-content\/uploads\/2023\/10\/80-Dynamic-Load-Balancing-100x100-1.png\" "\\"80-Dynamic-Load-Balancing-100×100\\"")
Dynamic Load Balancing<\/h6><\/div><\/div><\/div><\/div><\/div><\/div><\/div><\/div><\/div><\/div>
Spine Switch<\/h6><\/div>
AI\/ML Clusters<\/h6><\/div>
High-Performance Computing<\/h6><\/div>
Cloud DCI<\/h6><\/div>