QFX5240 Switch

Meet the foundation of your AI data center. The QFX5240 Switch leverages Broadcom Tomahawk 5 silicon to provide 800GbE—twice the capacity of the previous generation. It’s built for the advanced networking requirements of large-scale clusters, and works with Juniper Apstra automation to assure daily operation in AI and ML workload training and access. QFX5240 Switches deliver high-density 800GbE ports in a 2U fixed form factor with software to provide advanced network services tuned to the specific needs of AI/ML workloads. As a cornerstone of AI networks, these switches deliver fast job completion time (JCT) to accelerate training through high GPU utilization.

Key Features


Use Cases: AI data center leaf/spine, Data center fabric leaf/spine/super spine

Port Density:

  • QFX5240-64QD: 64 ports of QSFP-DD 800GbE. Can use breakout for 128 × 400GbE, or 256 × 100GbE
  • QFX5240-64OD: 64 ports of OSFP 800GbE. Can use breakout for 128 × 400GbE, 256 × 100GbE, or 256 x 50GbE

Throughput: Up to 102.4 Tbps (bidirectional)

Features + Benefits

Reduce space and total power use in your AI data center with 64 ports of 800GbE on a 2U switch—or use 2x400GbE for increased connectivity to GPUs—all while automating rail-optimized design through Apstra.

  • Choice of OSFP and QSFP-DD variants of 800GbE for leaf-spine connectivity
  • Advanced telemetry capabilities to support ECN/PFC counters
  • Fine-grained, load-balancing capability to handle reduced flow entropy

Junos OS Evolved Capabilities

The QFX5240 Switch supports advanced networking features such as BGP Additional Paths, MPLS, Layer 3 VPN, RDMA over Converged Ethernet (RoCEv2), and multicast.

Network Automation

Support for intent-based automation, Terraform, Python, Puppet, Ansible, and zero-touch provisioning (ZTP) addresses growing network complexity in Day 0 to Day 2+ operations.

Fabric Management

Intent-based Juniper Apstra software provides full Day 0 through Day 2+ capabilities for IP/EVPN fabrics with closed-loop assurance in the data center.

QFX Series Switches

See how the QFX5240 compares to other QFX products
QF5240-64OD Front with Top Low

QFX5240

The QFX5240 line offers up to 800GbE interfaces to support AI Data Center Networking deployments with AI/ML workloads and other high-speed, high-density, spine-and- leaf IP fabrics where scalability, performance and low latency are critical.

Technical Features

Use Cases: AI data center leaf/spine, Data center fabric leaf/spine/super spine

Port Density:

  • QFX5240-64QD: 64 ports of QSFP-DD 800GbE. Can use breakout for 128 × 400GbE, or 256 × 100GbE
  • QFX5240-64OD: 64 ports of OSFP 800GbE. Can use breakout for 128 × 400GbE, 256 × 100GbE, or 256 x 50GbE

Throughput: Up to 102.4 Tbps (bidirectional)

Currently viewing
QFX5230-64CD Front with Top Image View

QFX5230

The QFX5230 Switch offers up to 400GbE interfaces to support high-speed, high-density, spine-and-leaf IP fabrics. Prime use cases include AI data center networks with AI/ML workloads where scalability, performance, and low latency are critical.

Technical Features

Use Cases: Data center fabric spine, super spine (including AI data center networking), IP storage networking, and edge/data center interconnect (DCI)

Port density:

  • 64 ports of QSFP56-DD 400GbE. Can use breakout for 128 × 200GbE, 256 × 100GbE, 64 × 40GbE, 256 × 25GbE, or 256 × 10GbE

Throughput: Up to 25.6 Tbps (bidirectional)

qfx5220-32cd-frontwtop-low-1

QFX5220

The QFX5220 line offers up to 400GbE interfaces for very large, dense, and fast standards-based fabrics based on proven internet-scale technology. QFX5220 switches are an optimal choice for spine-and-leaf data center fabric deployments as well as metro use cases.

Technical Features

Use Case: Data Center Fabric Spine

Port Density:

  • QFX5220-32CD: 32 x 40/100/400GbE QSFP56-DD and 2 x 10GbE SFP+
  • QFX5220-128C: 128 x 100GbE QSFP28 and 2 x 10GbE SFP+

Throughput: Up to 25.6 Tbps (bidirectional)

 

Want to see every QFX product?

Try it. Right now.

Get hands-on with our IP-EVPN fabric solutions - for free!

Live Events and On-Demand Demos

Explore the journey to a transformed network.

Find QFX5240 in these solutions

AI Data Center

Deploy high performing AI training, inference, and storage clusters with speed and flexibility. Juniper Validated Designs offer confidence and speed, giving you up to 10x better reliability.

Data Center Networks

Simplify operations and assure reliability with the modern, automated data center. Juniper helps you automate and continuously validate the entire network lifecycle to ease design, deployment, and operations.

Data Center Interconnect

Juniper’s DCI solutions enable seamless interconnectivity that breaks through traditional scalability limitations, vendor lock-in, and interoperability challenges.

QFX5240 FAQs

What are the main use cases of the QFX5240 Switches?

The QFX5240 Switches are built for AI data center deployments as spine or leaf switches. They are also optimal for spine and super spine roles within any IP and EVPN-VXLAN fabrics.

What are the port densities and speeds of the QFX5240 Switches?

The QFX5240-64OD supports up to 64 x 800GbE, 128 x 400GbE, or 256 x 100GbE ports. The QFX5240-64QD supports up to 64 x 800GbE, 128 x 400GbE, 256 x 100GbE, or 256 x 50GbE ports.

What is the throughput of the QFX5240 Switches?

The QFX5240 switches deliver up to 102.4 Tbps bi-directional throughput with a forwarding capacity of 21.2 billion packets per second.