Overview
The Arista 7050X3 are members of the Arista 7050X series and key components of
the Arista portfolio of data center switches. The adoption of high performance
servers using virtualization and containers with increasingly higher bandwidth is
accelerating the need for dense 25 and 100G Ethernet switching in both the leaf and
spine tiers of modern networks. The Arista 7050X3 Series high performance exible
data center switches with a rich set of wire speed L2 and L3 features combine
extensive automation with programmability, low latency and consistent features for
software driven cloud networking.
The 7050X3 Series are available in a choice of conguration options. Combining
high density and industry leading power eciency with typical power consumption
under 7W per 100GbE port the 7050CX3-32S is ideal for both high performance leaf
or collapsed spine tiers with airow choices for back to front, or front to back.
Featuring 48 ports of 25G SFP ports and a choice of 8 or 12 ports of 100G ports the
7050SX3-48YC8 and 7050SX3-48YC12 enable high performance in a compact 1RU
form factor. The 7050SX3-96YC8 system has 96 ports of 25G and 8 ports of 100G in
2U. The 7050SX3 systems ensure seamless migration from 1/10G to 25G and enables
high density 25G solutions without network oversubscription. The 7050TX3-48C8
and 7050SX3-48C8 with 48 1/10G ports deliver consistent features and
enhancements for Enterprise and carrier networks.
All models of the 7050X3 Series oer exible forwarding tables with a Unied
Forwarding Table, latency from as low as 800ns and a fully shared packet buer of up
to 32MB for superior burst absorption. Comprehensive support for a wide range of
interface speeds including 10G, 25G, 40G, 50G and 100G combined with Arista EOS
ensures the 7050X3 delivers the exibility and features for big data, cloud, virtualized
and traditional network designs and accommodates the myriad dierent
applications and east-west trac patterns found in modern data centers.
7050X3 Series 10/25/40/50/100G Data Center Switches
Product Highlights
Performance
• 7050CX3-32S: 32x QSFP100
• 7050SX3-96YC8: 96x SFP25 & 8x QSFP100
• 7050SX3-48YC12: 48x SFP25 & 12x
QSFP100
• 7050SX3-48YC8: 48x SFP25 & 8x QSFP100
• 7050SX3-48C8: 48x SFP+ & 8x QSFP100
• 7050TX3-48C8: 48x 10G-T & 8x QSFP100
• Flexible 40G and 100G support
• Quad 10GbE and 25GbE support
• Up to 128 x 10G, 25G or 64x 50G
• Up to 6.4 terabits per second
• Up to 2 billion packets per second
• Wire speed L2 and L3 forwarding
• Latency from 800ns
Data Center Optimized Design
• 32 x 100G ports with under 7W / port
• Over 94% ecient power supplies
• 1+1 redundant & hot-swappable power
• N+1 redundant & hot-swappable fans
• Data center optimized airow
• Tool less rails for simple installation
Cloud Networking Ready
• VXLAN and VM Tracer
• OpenFlow, DirectFlow and eAPI
• 288K MAC entries and 360K IPv4 Routes
• 168K IPv4 Host Routes
• 32MB integrated intelligent buer with
dynamic buer allocation
Resilient Control Plane
• High Performance x86 CPU
• 8GB DRAM
• User applications can run in a VM
Advanced Provisioning & Monitoring
• CloudVision
• Zero Touch Provisioning (ZTP)
• LANZ for microburst detection
• DANZ Advanced Mirroring for visibility
• sFlow
• Self-congure and recover from USB
• Trac aware ECMP and UCMP
Arista Extensible Operating System
• Single binary image for all products
• Fine-grained truly modular network OS
• Stateful Fault Containment (SFC)
• Stateful Fault Repair (SFR)
• Full Access to Linux shell and tools
• Extensible platform - bash, python, C++,
GO, Opencong
Arista EOS
The Arista 7050X3 series runs the same Arista EOS software as all Arista products,
simplifying network administration. Arista EOS is a modular switch operating system
with a unique state sharing architecture that cleanly separates switch state from
protocol processing and application logic. Built on top of a standard Linux kernel, all
EOS processes run in their own protected memory space and exchange state
through an in-memory database. This multi-process state sharing architecture
provides the foundation for in-service-software updates and self-healing resiliency.
With Arista EOS, advanced monitoring and automation capabilities such as Zero
Touch Provisioning, VMTracer and Linux based tools can be run natively on the
switch with the powerful x86 CPU subsystem.
Arista 7050X3 Series Switches: 7050SX3-48YC12, 7050CX3-32S,
7050SX3-96YC8, 7050SX3-48YC8, 7050SX3-48C8 and 7050TX3-48C8
Arista 7050SX3-48YC12:
48x 25G SFP and 12x 100G QSFP ports
Arista 7050CX3-32S:
32x 100G QSFP100 ports, 2 SFP+ ports
7050X3 Series | Model Overview
The Arista 7050SX3-48YC12 is a 1RU system with 48 ports of 25GbE SFP and 12 ports of
100GbE QSFP with an overall throughput of 4.8Tbps. The high density SFP ports can be
congured in groups of 4 to run either at 25G or a mix of 10G/1G speeds. The QSFP ports
allow for a choice of 5 speeds including 100GbE, 40GbE, 4x10GbE, 4x25GbE or 2x 50GbE
with a wide choice of transceivers and cables enabling a choice of combinations for both
leaf and spine deployment. With low latency and no oversubscription, the switch is
optimized for high performance server and storage deployments.
High Availability
The Arista 7050X3 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)
• Multi-chassis LAG for active/active L2 multi-pathing
• 128-way ECMP routing for load balancing and redundancy
Model Overview
The Arista 7050X3 series come in dierent congurations. Each delivers high performance combined with feature rich layer 2 and layer
3 forwarding, suited for both top of rack leaf, or xed conguration spines.
7050SX3-48YC8, 7050SX3-48C8, 7050TX3-48C8 Rear View
7050CX3-32S 1RU Rear View (rear to front airow)
Arista 7050SX3-48YC8:
48x 25G SFP and 8x 100G QSFP ports
The Arista 7050SX3-48YC8 is a 1RU system with 48 ports of 25G SFP and 8 ports of 100G
QSFP with an overall throughput of 4Tbps. The high density SFP ports can be congured
to run either at 25G or a mix of 10G/1G speeds. The QSFP ports allow for 100GbE or 40GbE
as high speed network uplinks, with a wide choice of transceivers and cables enabling a
choice of combinations for both leaf and spine deployment. With low latency and no
oversubscription, the switch is optimized for high performance server and storage
deployments.
The 7050CX3-32S is a 1RU system with 32 100G QSFP ports oering wire speed
throughput of up to 6.4Tbps bi-directional. Each QSFP port supports a choice of 5 speeds
with exible conguration between 100GbE, 40GbE, 4x10GbE, 4x25GbE or 2x50GbE modes
for up to 128 ports of 10GbE and 25GbE or 64 ports of 50GbE. All ports can operate in any
supported mode without limitation, allowing easy migration from lower speeds and the
exibility for leaf or spine deployment.
7050SX3-48YC12 1RU Rear View (front to rear airow)
The Arista 7050SX3-48C8 and 7050TX3-48C8 are 1RU systems with 48 port of 10G and 8
ports of 100G with an overall throughput of 2.56Tbps. The SX3 high density 10G ports can
be congured for a mix of 10G/1G speeds to facilitate mixed connections and the TX3 RJ45
ports allows both 1G and 10G speeds. The QSFP ports allow 100GbE or 40GbE as network
uplinks, with a rich choice of transceivers and cables for both leaf and spine deployment.
With no oversubscription, the switch is optimized for high performance deployments with
consistent features.
Arista 7050TX3-48C8:
48x 10GBaseT and 8x 100G QSFP ports
The Arista 7050SX3-96YC8 is a 2RU system with 96 ports of 25G and 8 ports of 100G with
an overall throughput of 6.4Tbps bi-directional. The high density SFP ports can be
congured in groups of 4 to run either at 25G or a mix of 10G/1G speeds. The QSFP ports
allow for a choice of 5 speeds including 100GbE, 40GbE, 4x10GbE, 4x25GbE or 2x 50GbE
with a wide choice of transceivers and cables enabling a choice of combinations for both
leaf and spine deployment. With low latency and no oversubscription, the switch is
optimized for high density server environments with native 25G connections.
Arista 7050SX3-96YC8:
96x 25G SFP and 8x 100G QSFP ports
7050SX3-96YC8 2RU Rear View (front to rear airow)
7050X3 Series | Platform Overview
Dynamic Buer Allocation
In cut-through mode, the Arista 7050X3 switches forward packets with a consistent low latency of 800 nanoseconds. Upon congestion,
the packets are buered in an intelligent fully shared packet memory that has a total size of 32MB for superior burst absorption. Unlike
other architectures that have xed per-port packet memory, the 7050X3 Series use dynamic thresholds to allocate packet memory
based on trac class, queue depth and quality of service policy ensuring a fair allocation to all ports of both lossy and lossless classes.
Buer utilization, occupancy and thresholds are all visible with Arista LANZ and can be exported to monitoring tools to identify
hotspots and measure latency at the device and end to end.
Software Driven Cloud Networking
Arista Software Driven Cloud Networking (SDCN), combines the principles that have made cloud computing the unstoppable force
that it is: automation, self service provisioning, and linear scaling of both performance and economics coupled with the trend in
Software Dened Networking that delivers: network virtualization, custom programmability, simplied architectures, and lower capital
expenditure. This combination creates a best-in-class software foundation for maximizing the value of the network to both the
enterprise and service provider data center. A new architecture for the most mission-critical location within the IT infrastructure that
simplies management and provisioning, speeds up service delivery, lowers costs and creates opportunities for competitive
dierentiation, while putting control and visibility back in the hands of the network and systems administrators.
Maximum Flexibility for Scale Out Network Designs
Scale out network designs enable solutions to start small and evolve over time. A simple two-way design can grow as far as 128-way
without signicant changes to the architecture. The Arista 7050X3 include enhancements for exible scale-out designs:
•
128-way ECMP and 64-way MLAG to provide scalable designs and balance trac evenly across large scale 2 tier leaf-spine designs
•
Equal and Unequal Cost Multi-Pathing (UCMP) for exible trac balancing in large scale multi-tier topologies
•
Custom hash algorithms for ecient hashing, persistent hashing and custom lookups for tunneled protocols
•
Flexible allocation of L2 and L3 forwarding table resources for more design choice
•
Wide choice of dense 10G/25G/40G/50G/100G interfaces for multi-speed exibility
•
Support for standards based IEEE 25GbE for simple and cost eective migration from 10G and 40G to 25G and 100G
•
VXLAN routing, bridging and gateway capability for physical to virtualization communication in next generation data center designs
•
DANZ, sFlow and multi-port mirroring to detect micro-burst congestion and provide network wide visibility and monitoring
•
Hitless speed changes from 10G to 100G to eliminate down-time when implementing speed changes
Unied Forwarding Table
Network scalability is directly impacted by the size of a switches forwarding tables. In many systems a 'one size ts all' approach is
adopted using discrete xed size tables for each of the common types of forwarding entry. The Arista 7050X3 leverage a common
Unied Forwarding Table (UFT ) for the L2 MAC, L3 Routing, L3 Host and IP Multicast forwarding entries, which can be partitioned per
entry type. The ideal size of each partition varies depending on the network deployment scenario. The exibility of the UFT coupled
with the range of pre-dened proles available on the 7050X3 ensures optimal resource allocation for all network topologies and
network virtualization technologies.
Enhanced Features for High Performance Networks
The Arista 7050X3 series deliver a suite of advanced trac control and monitoring features to improve the agility of modern high
performance environments, with solutions for data monitoring, and next-generation virtualization.
Automating the data center enables customers to dynamically provision computing resources in the most ecient manner while also
meeting business needs by maintaining service level agreements (SLAs). Arista EOS automates complex IT workows and simplies
network operations while reducing or even eliminating downtime. Arista EOS rich automation capabilities not only reduce the human
error element in network operations but also enable IT operators to make the network work the way they want.
CloudVision
CloudVision is a network-wide approach for workload orchestration and workow automation as a turnkey solution for Cloud
Networking. CloudVision extends the EOS publish subscribe architectural approach across the network for state, topology, monitoring
and visibility. This enables enterprises to move to cloud-class automation without needing any signicant internal development.
7050X3 Series | Platform Overview
Advanced Event Management (AEM)
Simplifying the overall operations, AEM provides the tools to customize alerts and actions. AEM is a powerful and exible set of tools to
automate tasks and customize the behavior of EOS and the operation of the overall data center switching infrastructure. AEM allows
operators to fully utilize the intelligence within EOS to respond to real-time events, automate routine tasks, and automate actions based
on changing network conditions.
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 aect large numbers
of devices and cause signicant 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.
Virtualization
Supporting next-generation virtualized data centers requires tight integration with orchestration tools and encapsulation technologies
such as VXLAN. The 7050X3 build on the valuable tools already provided by the Arista VM Tracer suite to integrate directly into
encapsulated environments. Oering a wire-speed gateway between VXLAN and traditional L2/3 environments, they make integration
of non-VXLAN aware devices including servers, rewalls and load-balancers seamless and provide the ability to leverage VXLAN as a
standards based L2 extension technology for non-MPLS environments.
Precise Data Analysis
Arista Latency Analyzer (LANZ) is an integrated feature of EOS. LANZ provides precise real-time monitoring of micro-burst and
congestion events before they impact applications, with the ability to identify the sources and capture aected trac for analysis.
Advanced analytics are provided with features like buer monitoring with congurable thresholds, in-band path and latency
monitoring, event driven trace packets and granular time stamping.
Precision Timing (IEEE 1588)
Arista’s hardware derived Precision Time Protocol solution provides a robust mechanism for accurate in-band time distribution in high
performance environments. The system clock can be synchronized using IEEE 1588 PTP.
Dynamic Load Balancing *
Traditional hash-based load balancing algorithms can result in link and path allocations with short term imbalances and under
utilization of aggregate capacity. This is aggravated further in modern data centers with high trac loads, varied ow duration, mixed
packet sizes and micro-bursts. DLB enhancements to load balancing consider the real time load on links and dynamically assign new
and existing ows to the best link. When imbalances are detected active ows and new ows are allocated to the least loaded paths to
reduce the possibility of drops. Supported with any combination of ECMP and LAG/MLAG, DLB delivers higher throughout with
enhanced load distribution while oering the user an open implementation.
Flexible Pipeline
The Arista 7050X3 series support an enhanced forwarding architecture with smarter and exible packet pipeline which allows the
addition of new capabilities to the data plane of the packet processor through software upgrades without changes or replacement of
the underlying hardware. This allows for rapid testing and deployment avoiding costly replacements or major upgrades. Together with
exible resource allocation provided by the Unied Forwarding Tables (UFT), the programmable pipeline increases the exibility of the
platform allowing for broad use cases and ensures continued investment protection.
Network Address Translation
The Arista 7050X3 series support static and dynamic address translation at line rate and introducing no additional latency when the
mappings are set up. High performance environments can take advantage of NAT to resolve addressing challenges such as masking
internal addresses and translating overlapping ranges resulting in simpler network topologies without performance penalty.
*
Not currently supported in EOS