Arista 7060X5 (7060DX5-64S) 64x400GbE QSFP-DD AI/HPC Leaf Switch
High-density 400G leaf for RoCE-based AI and HPC fabrics
We help you choose, configure, and deliver the right system — no obligation.
Configuration at a Glance
Tailored per engagement. Full technical overview below.
Overview
The Arista 7060DX5-64S is a 2U fixed switch delivering 64 wire-speed 400GbE QSFP-DD ports at 25.6Tbps, configured for GPU-cluster leaf and high-radix HPC interconnect roles. Each unit is validated on the requested EOS build, tested across all 64 ports, and supplied through authorized Arista channels with manufacturer warranty intact.
Who This Solution Is For
Business Benefits
Massive 400G radix
Sixty-four 400G ports in 2U connect large GPU pods to spine without sprawling switch counts.
Tuned for AI traffic
Flow-aware scheduling and advanced congestion management keep RoCE and NVMe workloads lossless and predictable.
Low, consistent latency
Sub-microsecond cut-through forwarding with low jitter shortens collective-communication and all-reduce phases.
Typical Business Use Cases
GPU training cluster leaf switch
RoCEv2 lossless AI fabrics
High-radix HPC interconnect
NVMe-over-Fabrics storage aggregation
Industry Applications
Technical Overview
The 7060DX5-64S pairs a 25.6Tbps single-chip forwarding engine with 114MB of shared packet buffer and AI-optimized flow-aware scheduling for RoCEv2 and NVMe congestion control. Operating in cut-through mode it forwards at roughly 825 nanoseconds across all 64 QSFP-DD ports, with breakout to 256 lower-speed interfaces and full EOS and CloudVision integration.
| Switch Ports | 64 x 400GbE QSFP-DD (breakout to 256 x 100G) |
| Throughput | 25.6 Tbps (51.2 Tbps full-duplex) |
| Latency | ~825 ns cut-through, low-jitter |
| Packet Buffer | 114 MB shared packet buffer |
| AI Features | Flow-aware scheduling, RoCEv2 and NVMe congestion management |
| Form Factor | 2RU fixed, front-to-rear or rear-to-front airflow |
| Power | ~489W typical, dual redundant PSUs |
| Operating System | Arista EOS with CloudVision support |
| Warranty | Arista limited hardware warranty via authorized channel |
Specifications are indicative and configured to each engagement. Request a quote for a configuration tailored to your requirements.
Warranty, Support & Fulfillment
Every system ships from an authorized channel, configured and tested, with the documentation enterprise buyers need — backed by warranty and a dedicated account team.
Enterprise Warranty
Full manufacturer warranty with optional on-site, next-business-day support and extended coverage.
Authorized Channel
Sourced through Tier-1 distribution and OEM partners — never grey market. Asset & warranty records included.
Lead Time & Deployment
48-hour quotes, then configured, burn-in tested, and delivered on a committed schedule.
Nationwide Fulfillment
Coordinated logistics, rack-and-stack, and delivery wherever your infrastructure lives.
Frequently Asked Questions
How many GPUs can a 7060DX5-64S leaf support?
With 64 400G ports you can dedicate uplinks to spine while connecting dozens of 400G GPU-server NICs per rack; exact sizing depends on your rail design and oversubscription target, which we help validate during configuration.
Is this switch suitable for RoCEv2 AI training traffic?
Yes; flow-aware scheduling, advanced congestion management, and a 114MB buffer are tuned for lossless RoCEv2 and NVMe-over-Fabrics, which dominate GPU-cluster east-west traffic.
Should I use this as leaf or spine?
It excels as a 400G GPU-cluster leaf; for the largest lossless 800G AI spines the modular 7800R4 AI Spine is the matched upstream platform.
Hardware Assistance
Configure the Arista 7060X5 (7060DX5-64S) 64x400GbE QSFP-DD AI/HPC Leaf Switch with Nexus Compute
Tell us your requirements and a hardware specialist will help you specify, configure, and quote the right system — typically within two business days. No obligation.