rade with Higher Computing Performance

Bare-metal switches are driving the data center market as the high demands for cost-effective switch platfos will continue to be strong. hhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh

Overview

Bare-metal switches are driving the data center market as the high demands for cost-effective switch platforms and flexibility open hardware from service providers. With the upcoming era of 5G, the demands of Bare-metal boxes will continue to be strong. 5G advances the network applications which will need highly reliable and scalable infrastructure deployment.

800G Ethernet Switch for Data Denter and Cloud

Designed for AI clusters, hyperscale data centers, and high-performance computing environments, the TA064-IXM delivers industry-leading throughput and intelligent hardware architecture to meet the most demanding workloads of the future. QuantaMesh TA064-IXM is a high-performance and power efficient L2/3 switch designed in a 2RU form factor. QuantaMesh TA064-IXM is a next-generation 64-port 800G data center switch, designed to meet the scale and performance needs of modern AI clusters and hyperscale infrastructure. QuantaMesh TA064-IXM preloaded with ONIE, offering the flexibility to support a variety of network operating systems, including commercial 3rd party NOS, open-source platforms, and SONiC. The hardware architecture includes enhanced memory and storage capacity to support sophisticated automation tasks and management features. For enterprise-grade reliability, the switch is equipped with redundant power and cooling systems, featuring 1+1 hot-swappable power supply units and 3+1 hot-swappable fan modules to ensure high availability and continuous operation.

High Availability

The QuantaMesh BMS TA064-IXM provides high-performance solutions for cloud data centers and AI/ML applications. The key features from both hardware and software perspectives include:

• High capacity of 51.2T and high port density of 64x800G

• Handling the large, low entropy flows typical of AI/ML workloads

• Loaded with ONIE and capable of supporting multiple NOS, Commercial 3rd Party NOS, Open Source, SONiC

• Compact 2RU form factor

• Redundant PSUs and Fans