Juniper QFX5210-64C-AFO with 64 QSFP+/QSFP28 ports, reduced latency, redundant fans, 2 AC power supplies, front-to-back airflow
- Reviews (0)
The QFX5210 Switch is a flexible, 64-port, fixed-configuration switch designed for spine-and-leaf applications in next-generation IP fabric networks. Featuring line-rate, high-density QSFP+/QSFP28 ports supporting 10GbE, 25GbE, 40GbE, and 100GbE connections, the QFX5210 offers an advanced Layer 2, Layer 3, and MPLS feature set. This allows network operators to build large, high-density IP fabrics that support network virtualization and intelligent traffic forwarding based on proven, Internet-scale technology while enabling easy upgrades to 25GbE servers.
Network operators are increasingly deploying scale-out spine-and-leaf IP fabric architectures built with fixed-configuration switches to support growing east-west traffic in the data center. The Juniper Networks QFX5210 Switch is a next-generation, fixed-configuration spine/leaf switch that offers flexible, cost-effective, high-density 10GbE, 25GbE, 40GbE, and 100GbE interfaces for IP fabric networks, providing deployment versatility, investment protection, and future-proofing for today’s data centers. The QFX5210 Switch’s high port density also contributes to simplified Clos architectures, eliminating network tiers while reducing complexity and cost.
Like other Juniper Networks QFX Series Switches, the QFX5210 supports advanced L2, L3, and MPLS features. For large public cloud providers—among the first to adopt 25GbE servers to meet explosive workload growth—the QFX5210 enables very large, dense, and fast IP fabrics based on proven Internet-scale technology.
The QFX5210 runs the same reliable, high-performance Junos operating system used by network operators around the world.
Architecture and Key Components
The QFX5210 is a compact, 2 U platform featuring 64 quad small form-factor pluggable plus (QSFP+) transceivers or QSFP28 ports, with two redundant 1100 W AC/DC power supplies and four fan trays.
A QSFP28 module can be used to provide a 1x100GbE connection or 4x25GbE connection using breakout cables. A QSFP+ module can be used to provide a 1x40GbE connection or 4x10GbE connection using breakout cables.
The QFX5210 features a high-throughput 12.8 Tbps Packet Forwarding Engine (PFE). Control plane performance is further enhanced with the Intel Broadwell-DE, 4 Core, 2.2 GHz CPU, and 16 GB DDR4, 100 GB SSD storage.
The QFX5210 can be used in L3 fabrics and L2 networks with multi-chassis link aggregation (MC-LAG). Customers can choose the architecture that best suits their deployment needs and easily adapt and evolve as requirements change over time. The QFX5210 switch serves as a universal building block for these switching architectures, enabling data center operators to build cloud networks their way.
- IP Fabric: For customers looking to build scale-out data centers, an L3 spine-and-leaf Clos fabric is ideal due to its predictable, nonblocking performance and scale characteristics. For example, a two-tier fabric built with the QFX5210 Switch as leaves and QFX10000 Switches as the spine can scale to support up to 18,432 40GbE ports, or 36,864 25GbE or 10GbE server ports in a single fabric.
- Overlays: Customers can deploy overlay networks to provide L2 adjacencies for applications over L3 fabrics. The overlay networks use VXLAN in the data plane and EVPN to program the overlays. The overlays can operate without a controller or can be orchestrated with a fabric management orchestrator to provide L2/L3 overlay virtual networking and security for bare-metal servers and virtual workloads. QFX5210 switches can also integrate with VMware NSX.
- MC-LAG: The QFX5210 supports the MC-LAG protocol between two switches, eliminating Spanning Tree Protocol (STP) in traditional L2 networks when deployed in the aggregation layer. MC-LAG’s active/active operation ensures complete bandwidth utilization between the network’s access and aggregation layers, while the dual control plane technology ensures the highest availability for applications.