7260X3 Series | Model Overview

The 7260CX3-64E retains all features of the 7260CX3-64 and additionally provides an enhanced power and cooling capability, support

ing the use of 4.5W power rated optics on all ports, when used with AC or DC 1900W power supplies and high speed fans.

Consistent features to both the Arista 7060CX and 7260CX Series combined with lower power and high 100GbE density means the

7260CX3 is optimized for 100GbE top of rack and spine tiers, high density storage and next generation nancial trading systems

requiring predictable performance with low latency.

Dynamic Bu er Allocation

The Arista 7260CX3 switches o er consistent latency from 450ns in cut-through mode, and a shared 42 MB integrated packet bu er

that is twice the capacity of the 7060X Series. Upon congestion from micro-bursts or fan-in packets are bu ered in an intelligent shared

packet memory that has a total size of 42MB arranged as 10.5MB per port group to accommodate bursts and lossless tra c

requirements. Unlike other architectures that have xed per-port packet memory, or smaller shared memory pools the 7260CX3 Series

use a Dynamic Bu er Allocation (DBA) scheme to allocate memory intelligently based on a combination of tra c class, queue depth

and quality of service policy ensuring fair allocation to all ports. Bu er utilization, occupancy and thresholds are all visible with Arista

LANZ and can be exported to monitoring tools for detailed analysis.

Model Overview

The 7260CX3-64 is a 2RU system with 64 100G QSFP ports

o ering wire speed performance with an overall throughput of up

to 12.8 Tbps. Each QSFP port is capable of a choice of 100GbE,

40GbE, 4x10GbE, 4x25GbE or 2x50GbE with hitless con guration

change between modes. The 64 QSFP ports can be broken out to

a system maximum of 128 ports allowing for easy transitions and

maximum exibility enabling deployment as both a leaf and spine.

The two SFP+ ports provide additional connections for low speed

management networks or out of band monitoring.

AI Analyzer *

Traditional software-based tra c counters do not lend themselves to examine AI/ML tra c patterns, which exhibit unique ramp up

behavior in very short intervals of time. The AI Analyzer is a hardware capability that enables the collection of ECMP member utilization

data, aggregated over extremely short periods of time. This allows the Arista 7260X3 series to e ectively analyze the tra c patterns,

with a time interval as granular as 100 microseconds. The results of such an analysis can then be applied to ne tune dynamic load

balancing workloads uniformly across the ECMP member links, which is a key requirement for AI/ML applications.

Smart System Upgrade

Smart System Upgrade is a network application designed to address one of the most complicated and challenging tasks facing data

center administrators - network infrastructure maintenance. Changes to the underlying network infrastructure can a ect large numbers

of devices and cause signi cant outages. SSU provides a fully customizable suite of features that tightly couples data center

infrastructure to technology partners allowing for intelligent insertion and removal, programmable updates to software releases and

open integration with application and infrastructure elements.

High Availability

The Arista 7260X3 series switches are designed for high availability

from both a software and hardware perspective. Key high availability

features include:

• 1+1 hot-swappable power supplies and four N+1 hot-swap fans

• Color coded PSU’s and fans

• Live software patching

• Self healing software with Stateful Fault Repair (SFR)

• Smart System Upgrade (SSU) and Accelerated Software Update (ASU)

Arista 7260X3 2RU Rear View: Front to Rear air ow (red)

Arista 7260X3 2RU Rear View: Rear to Front air ow (blue)

Arista 7260CX3-64:

64x 100GbE QSFP100 ports, 2 SFP+ ports

* Not currently supported in EOS