Brocade 6520 Data Center Switches

btn_contact btn_download

The Brocade 6520 is a 96-port, high-density switch for large enterprise data centers to support data center consolidation, growing workloads, and highly virtualized, private cloud storage environments. Delivering market-leading Gen 5 Fibre Channel technology and unmatched port density, the Brocade 6520 Switch provides industry-leading scalability, reliability, and 16 Gbps performance in a flexible, easy-to-deploy enterprise-class switch.

Class-Leading Performance for Today and Tomorrow

Brocade-6520-Data-Center-SwitchThe Brocade 6520 Switch meets the demands of growing, dynamic workloads and private cloud storage environments by delivering market-leading Gen 5 Fibre Channel technology and capabilities. The Brocade 6520 is a high-density, purpose built, foundational building block for large and growing Storage Area network (SAN) infrastructures. It provides industry-leading scalability, reliability, and 16 Gbps performance in a flexible, easy-to-deploy enterprise-class switch, enabling greater data center consolidation, operational efficiency, and business continuity. In addition to increased throughput, it helps improve bandwidth utilization, security, and network visibility and management through in-flight data compression and encryption and advanced diagnostics. It’s an ideal switch for bandwidth-intensive workloads, evolving virtualized data centers, and private cloud architectures.

Additional highlights include the following:

  • Provides high scalability in an ultra-dense 96-port switch to support highly virtualized, private cloud storage and data center consolidation
  • Enables “pay-as-you-grow” flexibility from 48 to 96 ports using Ports on Demand (PoD) capabilities with speeds up to 16 Gbps
  • Simplifies management through Brocade Fabric Vision technology, reducing operational costs and improving application performance
  • Helps maximize application uptime and performance while reducing overall operational expenses with ClearLinkDiagnostic Ports (D_Ports)
  • Provides up to eight in-flight encryption and compression ports, delivering data center-to-data center security and bandwidth savings
  • Optimizes link and bandwidth utilization with Brocade ISL Trunking and Dynamic Path Selection (DPS)
  • Accelerates deployment and troubleshooting with Dynamic Fabric Provisioning (DFP), monitoring, and advanced diagnostics

Dynamic Path Selection (DPS) augments Brocade ISL Trunking to route data efficiently between multiple trunk groups.

Features and Benefits

Exceptional Scalability for Demanding Workloads and Data Center Consolidation The Brocade 6520 features 96 Fibre Channel ports in a 2U form factor, delivering industry-leading port density and space utilization for data center consolidation. Designed for maximum flexibility, this enterprise-class switch offers “pay-as-you-grow” scalability with Ports on Demand (PoD). Organizations can quickly, easily, and cost-effectively scale from 48 to 96 ports in 24-port increments, each supporting 2, 4, 8, 10, or 16 Gbps. In addition, flexible, high-speed 16 Gbps and 8 Gbps optics allow organizations to deploy bandwidth on demand to meet growing data center needs. For maximum flexibility, the switch also features dual-direction airflow options to support the latest hot aisle/cold aisle configurations.
Industry-Leading Performance for Growing Workload The Brocade 6520 delivers exceptional performance for growing and dynamic workloads through a combination of market-leading throughput and bandwidth utilization. With the unpredictability of virtualized workloads and cloud services, throughput becomes critical to ensuring that the network does not become the bottleneck. With 96 ports, the Brocade 6520 provides an aggregate 1536 Gbps full-duplex throughput. Up to eight ISLs can be combined together in a 128 Gbps framed-based trunk. In addition, exchange-based Dynamic Path Selection (DPS) optimizes fabric-wide performance and load balancing by automatically routing data to the most efficient, available path in the fabric. This augments Brocade ISL Trunking to provide more effective load balancing in certain configurations. Moreover, the enterprise-class capabilities of this switch yield 40 percent higher performance compared to 10 Gigabit Ethernet (GbE) alternatives at a similar cost.
Simplified Management and Robust Network Analytics Brocade Fabric Vision technology, an extension of Brocade Gen 5 Fibre Channel, introduces a breakthrough hardware and software technology that maximizes uptime, simplifies SAN management, and provides unprecedented visibility and insight across the storage network. Offering innovative diagnostic, monitoring, and management capabilities, the Brocade 6520 with Fabric Vision technology helps administrators avoid problems, maximize application performance, and reduce operational costs. Brocade Fabric Vision technology includes:

  • Brocade ClearLink diagnostics
  • Bottleneck Detection
  • Integration into Brocade Network Advisor
  • Forward Error Correction (FEC)
  • Additional buffers
  • Real-time bandwidth consumption by hosts/applications on ISLs
A Building Block for Virtualized, Private Cloud Storage The Brocade 6520 provides a critical building block for today’s highly virtualized, private cloud storage environments. It simplifies server virtualization and Virtual Desktop Infrastructure (VDI) management while meeting the high-throughput demands of Solid State Disks (SSDs). The Brocade 6520 also supports multitenancy in cloud environments through Virtual Fabrics, Quality of Service (QoS), and fabric-based zoning features.

The Brocade 6520 enables secure metro extension to virtual private or hybrid clouds with 10 Gbps Dense Wavelength Division Multiplexing (DWDM) link support, as well as in-flight encryption and data compression to optimize bandwidth and minimize the risk of unauthorized access. With four times more in-flight encryption and compression ports than the Brocade 6510 Switch, the Brocade 6520 supports higher data volumes over long distance. The switch also features on-board data security and acceleration, minimizing the need for separate acceleration appliances to support distance extension. Internal fault-tolerant and enterprise-class RAS features help minimize downtime to support mission-critical cloud environments.

To learn more about PTS’ IT Consulting Services, visit a link below or contact us:



Arista Networks 7500 Series Data Center Switch

btn_contact btn_video

The Arista 7500 Series modular switch sets a new standard for performance, density, reliability, and power efficiency for data center switches. In a compact 7RU (4-slot) or 11RU (8-slot) chassis, the Arista 7500 Series offers up to 10 Terabits of fabric capacity, 384 wirespeed 10 GbE ports and 5.76 BPPS of L2/3 throughput. The Arista 7500 offers a comprehensive feature set, as well as future support for the 40GbE and 100GbE standards. With front-to-rear airflow, redundant and hot swappable supervisor, power, fabric and cooling modules, the 7500 is energy efficient with typical power consumption of 10 watts per port for a fully loaded chassis. All of these attributes make the Arista 7500 an ideal platform for building low latency, scalable, data center networks.

Arista Networks 7504 and 7508 Modular Data Center Switches

Arista EOS

All Arista products including the 7500 series switches run Arista EOS software. The same binary image supports all Arista products, making it easy for network administrators to standardize the operating system across all switches in the data center. With the addition of Arista 7500, EOS has been extended to support redundant supervisor modules with stateful switchover without the loss of data plane forwarding.

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 a memory based database. This multi-process state sharing architecture provides the foundation for in-service-software updates and self-healing resiliency. Arista EOS automates the networking infrastructure natively with VMware vSphere via VM Tracer to provide VM discovery, auto-VLAN provisioning, and unprecedented visibility into the virtual computing environment.

Additional highlights include the following:

Modular 10Gb Ethernet Switching

  • Offered in 8 and 4-slot configurations
  • 384 or 192 wirespeed 1/10Gb Ethernet port options
  • Up to 5.76Bpps L2 and L3 switching and routing
  • 2.3GB packet buffer per linecard module
  • 4.5usec port to port latency

System Capabilities and Architecture

  • Up to 10 Terabit lossless switch fabric
  • 648Gb/s bandwidth per linecard Slot (1.25 Terabit full-duplex)
  • Low-latency, lossless, VOQ fabric
  • Data Center Class airflow and resiliency

Advanced Technology

  • Designed for real-world deployments of 40Gb and 100GbE
  • Data Center Bridging – lossless Ethernet transport

Energy Efficiency

  • Less than 10 watts per wirespeed 10GbE port
  • Gold Climate Saver power supplies
  • Front-to-rear cooling and HVAC
Arista Simple and Scalable Layer-2 Leaf-Spine Architecture
Arista Simple and Scalable Layer-2 Leaf-Spine Architecture

Scaling Data Center Performance

The Arista 7500 enables dramatically faster and simpler network designs for large-scale data centers. When used in conjunction with Arista 7048 or 7100 leaf switches and Arista’s Multi-Chassis Link Aggregation (MLAG) technology, a pair of 7508 Switches can support 9200 Servers with a leaf/spine active/active L2 network topology. With four Arista 7508 switches at Layer 3, more than 18,000 Servers can be connected with a non-blocking, low-latency, two-stage network that provides predictable end-to-end application performance.

Deterministic Network Performance

The Arista 7500 uses a deep buffer virtual output queue (VOQ) architecture that eliminates head-of-line (HOL) blocking and virtually eliminates packet drops even in the most congested network scenarios. An advanced traffic scheduler fairly allocates bandwidth between all virtual output queues while accurately following queue disciplines including weighted fair queuing, fixed priority, or hybrid schemes including 802.1Qaz ETS. As a result, the Arista 7500 can handle the most demanding data center requirements with ease, including mixed traffic loads of real-time, multicast, and storage traffic.

Designed for High Availability

The Arista 7500 is designed specifically for high-availability requirements. The hardware supports hot-swappable redundant power, cooling, fabric, supervisors, and linecards. The fabric is N+1 redundant with graceful degradation. The Arista EOS software supports stateful failover between the redundant supervisors as well as self healing and live patching through in-service- software updates.


7500 Chassis – 8-slot and 4-slot

The 7500 chassis provides room for two supervisor modules, four or eight linecard modules, four power supply modules, and six fabric modules. The 7504 chassis fits into 7 rack units while the 7508 chassis fits into 11U of a standard data center rack. Supervisor and linecard modules plug in from the front, while the fan, fabric and power supply modules are inserted from the rear. The midplane is completely passive and provides control plane connectivity to each of the fabric and linecard modules. The system design is optimized for data center deployments with front-to-rear airflow.

7500 Supervisor Module

The supervisor modules for the 7500 series run Arista Extensible Operating System (EOS) and handle all control plane and management functions of the system. One supervisor module is needed to run the system and a second can be added for 1+1 control plane redundancy. Each supervisor module takes up only a half slot resulting in very efficient use of space and a higher density design. The dual-core x86 CPU provides the necessary control plane performance needed to run an advanced data center switch. Each supervisor module also includes an LCD panel in the front to monitor system status making it easy for network operators to quickly identify issues.

7500 Fabric Module

At the heart of the 7500 series is the fabric. It interconnects all linecards in a non-blocking architecture irrespective of the traffic pattern. Each linecard module connects to the fabric with multiple links and data packets are spread across these links to fully utilize the fabric capacity using a fabric control protocol. Unlike typical hash-based selection of fabric links, the 7500 architecture provides connectivity from any port to any other port with no drops. The fabric modules are N+1 redundant and can be hot swapped with very little impact to data traffic. The Fabric Module for the 7508 Chassis and the 7504 Chassis are different based on the size of the chassis and the 7504 Fabric integrates the fan assembly.

7500 Power Supply Module

The 7500 series switches are provisioned with four 2900W AC power supplies. The power supplies are 2+2 grid redundant and hotswappable. Each power supply must be connected to a 16A (20A UL) 200-240V AC circuit. The power supplies are gold climate saver rated and have an efficiency of over 90% with single stage conversion to the internal 12V DC voltage.

7548S Linecard Module

The 7548S Linecard offers 48 wirespeed 1/10GbE SFP+ ports with an aggregate throughput of 760 Mpps at both Layer 2 and Layer 3. The Arista 7508 chassis can be fully populated with eight linecards to achieve a wirespeed 384 10GbE ports. The virtual output queue architecture eliminates head-of-line blocking and with 2.3GB of packet memory on each linecard, approximately 40 msec of traffic can be buffered per ingress port, virtually eliminating packet drops in congestion scenarios. Each linecard module connects to all fabric modules in a non-blocking full mesh. The redundant fabric modules allow the system to continue forwarding.

To learn more about PTS’ IT Consulting Services, contact us or visit:

lt_arrowBack to Arista Networks Manufacturer Page

Arista Networks 7100 Series Data Center Switch


Arista-Networks-7100-Series-Data-Center-SwitchThe Arista 7100 Series of Datacenter Ethernet switches feature the industry’s highest density 10 Gigabit Ethernet switching solution and the first with an extensible modular network operating system. With breakthrough price-performance, the Arista 7100 Series enables 10 Gigabit Ethernet to be deployed everywhere in the data center, which can significantly improve server utilization and consequently data center power efficiency.

Additional highlights include the following:

High-Performance, Low Latency 10 Gigabit Ethernet Switching

  • Layer 2/3/4 switching up to 960 Gbps and 720 Mpps
  • As low as 500 nanosecond latency at all packet sizes

Arista EOS™

  • Extensible operating system customizable to customer needs
  • Fine-grain modular protected memory architecture
  • Access to Linux tools

High Availability

  • In-service-software-upgrades (ISSU)
  • Self-healing stateful fault repair (SFR)
  • Hot swappable, redundant power supplies and fans

Deployment Flexibility

  • All ports support a wide selection of 10GbE or GbE PHYs
  • Front-to-rear and reversible airflow options for flexible mounting

Datacenter Optimized

  • 1 rack unit (RU) for 24 or 48-port models
  • Power efficient

Cloud Leaf and Spine Network Design

Arista-Networks-7100-Series-Network-DesignIn the data center, there is an architectural migration to two-tiered cloud networking designs. The main building blocks are Cloud Leaves (CL) and Cloud Spines (CS). Cloud Spines forward traffic along optimal paths between nodes at layer 2 or layer 3 while Cloud Leaves control the flow of traffic between servers.

Predictable Latency and Multicast

The ultra-low latency characteristics of the Arista 7100 Series are consistent regardless of packet size and port speed, which means customers can reap the benefits of industry leading latency at 1 Gigabit speeds in addition to 10 Gigabit speeds. The 7100 Series latency is also predictable with little variance in environments with random packet sizes which is key to improving and maintaining application performance and server utilization.

Multicast and broadcast throughput is also very consistent at all packet sizes, maintaining line rate performance with no packet drops. This means that multimedia applications such as content delivery and video conferencing can be used in the network in a reliable and effective way.

High Availability

The Arista 7100 Series was designed for high availability from both a software and hardware perspective. Key high availability features include:

  • Two 1+1 hot-swappable independent power supplies
  • Five N+1 hot-swappable independent fans
  • Dual management ports
  • In-Service Software Upgrades (ISSU)
  • Self healing software with Stateful Fault Repair (SFR)
  • Up to 16 10GbE ports per link aggregation group (LAG)
  • ECMP routing for load balancing and redundant fail over

Target Positioning

The Arista 7100S Switches are positioned for a variety of places in the network including:

  • Top-of-Rack Server Aggregation
  • Blade Server Aggregation
  • Network Spine and Aggregation
  • HPC Cluster Interconnects
  • Storage Access
  • Cloud Networking Interconnects

10 Gigabit Ethernet is becoming the interconnect of choice in data centers of all sizes providing key improvements in terms of bandwidth, latency, scalability, reliability and application performance. The Arista 7100 Series enables a seamless migration path from existing Gigabit Ethernet-based servers to 10 Gigabit Ethernet-based high-performance servers while further enabling the transition to virtualized environments.

Model Comparison


To learn more about PTS’ IT Consulting Services, contact us or visit:

lt_arrowBack to Arista Networks Manufacturer Page

Arista Network 7050 Series – 10/40 GbE Data Center Switch


Arista Networks 7050 Series Switch

The Arista 7050S-64 switch is a wire speed layer 2/3/4 switch with 48 10GbE SFP+ and 4 40GbE QSFP+ ports in a compact 1RU chassis. Each 40GbE port can also operate as four independent 10GbE ports to provide a total of 64 10GbE ports. The Arista 7050-64 offers a latency of 800 to 1200 ns in cut-through mode, and a shared 9 MB packet buffer pool is allocated dynamically to ports that are congested. With a typical power consumption of less than 2 Watt/port with twinax copper cables, and less than 3 Watt/port with SFP/QSFP lasers, the 7050S-64 provides industry leading power efficiency for the data center. An optional built-in SSD supports advanced logging, data captures and other services directly on the switch.

Additional highlights include the following:


  • 48×1/10GbE SFP and 4x40GbE QSFP ports
  • 1.28 terabits per second
  • 960 million packets per second
  • Wire speed L2 and L3 forwarding

Data Center Optimized Design

  • 125W typical power draw
  • 1+1 redundant & hot-swappable power
  • N+1 redundant & hot-swappable fans
  • Front-to-rear or rear-to-front cooling

Cloud Networking Ready

  • 128K MAC entries
  • TRILL support
  • 16K IPv4 Routes
  • 16K IPv4 Host Routes
  • 8K IPv6 Routes
  • 8K Multicast Groups
  • 9MB Dynamic Buffer Allocation

Resilient Control Plane

  • Dual-core x86 CPU
  • 4GB DRAM
  • 2GB Flash
  • User applications can run in a VM

Built-in Storage

  • Solid State Drive option
  • Store logs and data captures
  • Network boot nodes from the switch
  • Linux tools with no limitations

Advanced Provisioning & Monitoring

  • Zero Touch Provisioning (ZTP)
  • VMTracer – host view
  • VMTracer – auto vlans
  • sFlow
  • Self-configure and recover from USB

Arista Extensible Operating System

  • Single binary image for all products
  • Fine-grained modularity
  • Stateful Fault Containment (SFC)
  • Stateful Fault Repair (SFR)
  • Access to Linux tools
  • Extensible platform

High Availability

The Arista 7050S-64 switch was designed for high availability from both a software and hardware perspective.  Key high availability features include:

  • 1+1 hot-swappable power supplies
  • Four N+1 hot-swappable fans
  • Live software patching
  • Self healing software with Stateful Fault Repair (SFR)
  • Up to 16 10GbE ports per link aggregation group (LAG)
  • Multi-chassis LAG for active/active L2 multipathing
  • 32-way ECMP routing for load balancing and redundancy
  • Arista 7050S-64: Latency using RFC 2544 tests

Dynamic Buffer Allocation

In cut-through mode, the Arista 7050S-64 forwards packets with a latency of 800 to 1200 nanoseconds. Upon congestion, the packets are buffered in shared packet memory that has a total size of 9 MB. Unlike other architectures that have fixed per port packet memory, the 7050S-64 uses Dynamic Buffer Allocation (DBA) to allocate up to 5MB of packet memory to a single port for lossless forwarding.

Arista 7050S-64: Latency using RFC 2544 tests

Cloud Network Scale: Up to 18K 10GbE nodes using standardized protocols



To learn more about PTS’ IT Consulting Services, contact us or visit:

lt_arrowBack to Arista Networks Manufacturer Page


Free Report:
Data Center
Industry Trends
close slider

Download FREE Data Center Industry & IT Infrastructure Trends PDF Now!

PTS Data Center Solutions Company Overview with Industry Trends Data