Arista 7130L Series FPGA-enabled switches
The 7130L FPGA-enabled devices leverage FPGA technology to enable the development and deployment of cutting-edge network applications. Available in 32, 48 or 96 SFP+ port options, the 7130L is multiple devices in one; performing layer 1+ switching in only 5 ns, enabling unrestricted access to an onboard FPGA and containing an x86_64 server.
The Arista 7130L Series is optimised for Arista’s MetaMux and MetaWatch network application and can equally be leveraged to run 3rd party partner applications. FPGA application developers can utilise the platform to deploy and deliver their performance critical apps. On top of the market-leading FPGA functionality, the devices combine a multitude of Layer 1+ network functionality on to the same devices:
- Signal regeneration
- Port mirroring
- Dynamic patching/link management
- Ad-hoc tapping without rewiring
- Layer 1+ statistics on every link
- Media conversion
- Telemetry and more
The devices provide layer 1+ configurable features such as dynamic patching, tapping, one-to-many replication, media conversion, packet stats and precise timestamping. The also include the following functionality:
- 5 ns layer 1 switching between front panel ports
- 3 ns latency from front panel to FPGA
- Various specifications for RAM and Transceivers
- Development toolkits and low-latency IP Cores
Arista 7130L Technical Specifications
Model Comparison of 7130L Series devices
7130L Series Devices | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Model |
FPGA |
FPGA Quantity |
SFP+ Ports |
FPGA Ports |
Off-Chip RAM |
RU |
ePCIE |
PPS In/Out |
Clock |
SSD Bays |
Internal 10G Ports |
48L | Xilinx Virtex® UltraScale+™ VU7P | 1 | 48 | 60 | 4x 8GB DDR4 2400 ECC | 1 | x | OCXO | x | ||
96L | Xilinx Virtex® UltraScale+™ VU7P | 1 | 96 | 58 | 4x 8GB DDR4 2400 ECC | 2 | x | OCXO | x | ||
32LB | Xilinx Virtex® UltraScale+™ VU9P | 1 | 32 | 60 | 4x 8GB DDR4 2400 ECC | 1 | x | x | OCXO | x | x |
48LB | Xilinx Virtex® UltraScale+™ VU9P | 1 | 48 | 60 | 4x 8GB DDR4 2400 ECC | 1 | x | OCXO | x | ||
96LB | Xilinx Virtex® UltraScale+™ VU9P | 1 | 96 | 58 | 4x 8GB DDR4 2400 ECC | 2 | x | OCXO | x | ||
48LA | Xilinx Virtex® UltraScale+™ VU7P | 1 | 48 | 60 | 4x 8GB DDR4 2400 ECC | 1 | x | Atomic | x | ||
96LA | Xilinx Virtex® UltraScale+™ VU7P | 1 | 96 | 58 | 4x 8GB DDR4 2400 ECC | 2 | x | Atomic | x | ||
48LBA | Xilinx Virtex® UltraScale+™ VU9P | 1 | 48 | 60 | 4x 8GB DDR4 2400 ECC | 1 | x | Atomic | x |
Robust switches with comprehensive feature sets
Feature | Description |
---|---|
Simplified stack | Fan-out with 5 ns of latency, equivalent to a single metre of fiber or copper interconnect, and insignificant jitter. |
Integrated with existing FPGA tools | Reduce costs by converting between different Ethernet media types e.g. one end of a link can be twinax and the other 10GBASE-LR single-mode fiber. |
Layer 1+ functionality | Save rack space and reduce complexity by leveraging dynamic patching, tapping, one-to-many replication, telemetry and comprehensive port statistics on a single device. |
Feature rich | Avoid the need to build features in-house by leveraging Arista’s access control, syslog, SNMP, packet stats, tcpdump, JSON RPC API, time series data, streaming telemetry and more - included as standard within the 7130 Series. |
Easy app deployment | Streamline operational processes through Arista’s built-in application infrastructure which allows developers to wrap applications into simple packages for deployment. |
Enterprise ready | Deploy FPGA applications with ease - the FPGA platform integrates with a 64-bit x86 management processor and the MOS Operating system to provide user extensible solutions. |
Ultra-low latency | Fan-out with 5 ns of latency, equivalent to a single metre of fiber or copper interconnect, and aggregate in 39 ns. |
MOS Operating System
MOS provides a core set of features that are common across the 7130 platform. It is based on Linux and provides a command line and web interface as well as support for other management protocols. MOS provides a standard, mature and powerful platform with the commands, tools and packages such as syslog, net-snmp, daemons, RPMs, Bash, Python, authentication, and security.
MetaWatch
MetaWatch is a powerful network application designed to simplify tapping networks. It enables traffic capture with high-resolution timestamping and allows for advanced network monitoring and detailed network analytics. MetaWatch combines several components of a traditional network monitoring solution into one powerful device:
- Tapping with negligible latency impact
- Flow-control of aggregated captured traffic
- Time synchronization
- Subnanosecond-precise timestamping
- Deep buffering
- Multi-port capture
MetaWatch provides all features with virtually no impact on the monitored network performance and enables a seamless stream of time stamped frames to storage and analytics tools.
MetaMux
MetaMux is a ultra-low latency network application designed for Arista's 7130E and L Series switches performing multiplexing/aggregation of incoming streams in an average of 39 ns*.
The MetaMux application provides a configurable number of multiplexer (mux) options all the way from 48:1 down to multiple 4:1 mux instances. To allow for the situation where multiple packets arrive at the multiplexer at the same time, Metamux will queue packets in its input buffers to ensure contending packets can be sent on the link. Buffer utilisation and statistics are available from the device. For more information visit the Applications page.
* Average latency on 7130LB devices
ExchangeApp
Arista ExchangeApp is a network application that performs highly accurate, low-latency, inline packet timestamping. By providing exchange software systems with the precise arrival times of trading orders, ExchangeApp makes it possible to build fairer financial markets.
ExchangeApp applies an inline timestamp at the exchange edge and appends it to each packet before that packet is forwarded out of a low latency interface. On this low-latency passthrough path, packets are received, timestamped and output with < 200ns.
MultiAccess
MultiAccess is a network application that combines low-latency packet multiplexing with the ability to segregate traffic between individual clients, or groups of clients, via configurable filters.
Arista MultiAccess enables the creation of a shared, low-latency infrastructure ideal for managed service providers or brokers providing direct access to clients .
The MultiAccess application optimizes the paths for many-to-one environments where it is critical that clients maintain segregation of traffic between each endpoint. These paths are further enhanced to allow the processing to occur at the fraction of the latency of a Layer 2/3 switch.
MetaProtect
MetaProtect Firewall is a network application that runs on the 7130L Series devices performing line-rate low-latency parallel packet filtering and logging between port-pairs. Filtering is implemented via per-port Access Control Lists (ACL). MetaProtect Firewall provides complete flexibility in configuration, allowing authenticated administrators to create mappings between physical port-pairs and apply either uni or bidirectional security policy
SwitchApp
Arista SwitchApp is a full-featured 1/10/40G Layer 2 switching app implemented in FPGA, optimised for low latency and integrated with EOS.
Data Sheet
Product Overview
White Papers
- .7130 a Better Way to Tap
- .5 Ways to Optimise exchange connectivity latency
- .5 Things to Consider When Choosing an FPGA platform
- .Measuring the Absolute Accuracy of 10GbE Packet Timestamping
- .Four key trends in the networked use of FPGAs
- .STAC-TS™ BENCHMARKS
Accuracy of network timestamping and burst capacity of capture
Product Briefs
- .MetaWatch Product Brief
- .MetaMux Product Brief
- .ExchangeApp Product Brief
- .MultiAccess Product Brief
- .SwitchApp Product Brief
- .MetaProtect Product Brief
Solution Briefs
- .An Overview of Arista Ethernet Capture Timestamps
- .FPGA Platforms: Why the Arista 7130 is best of breed
- .Network Traffic Capture & Aggregation: Why buffer size is crucial
- .Improving MetaWatch’s timestamp accuracy with clock modules
- .Copper is Faster than fiber