Nokia 7250 IXR

The Nokia 7250 IXR uses merchant silicon from the Broadcom Jericho/Qumran family, positioning it as a low-latency, high-throughput aggregation and data center edge platform.

Field Value
Total Buffer Moderate (shared pool, varies by ASIC)
Buffer Model Shared + VOQ (Virtual Output Queues)
Silicon Broadcom Jericho/Qumran family
Target Interfaces 100G / 400G
Target Role Aggregation, DC edge

Buffer Architecture

VOQ (Virtual Output Queue)

Packets are queued per egress destination rather than per ingress port. This eliminates head-of-line blocking — a slow or congested output port cannot stall traffic destined for a healthy port.

Shared Buffer Pool

Memory is dynamically allocated across ports. The total pool is smaller than FP-based systems (Nokia 7750 SR) but is used efficiently through statistical multiplexing.

Depth Profile

Shallow-to-moderate by design. The 7250 IXR prioritizes low latency and high throughput over sustained burst absorption.

What It Handles Well

Limitations vs FP-based platforms

See the Nokia Access/Aggregation architecture page for cross-platform comparison. Additionally, you can read the Broadcom Qumran architecture page for deep-dive details on the chipset.

← Back to the main buffer table