Calnex SNE Ignite FPGA-Based Precision Network Emulator for O-RAN, 5G and DCI

FPGA-based multi-port network emulator with 6.8 µs intrinsic latency and line-rate impairments from 10GbE to 400GbE.

Optical interfaces – 400GbE: QSFP-DD (DR4/FR4/LR4/SR4/SR8/LR8/ER8) – 4 ports, 1 bidirectional path
Optical interfaces – 100GbE: QSFP28 (LR4/SR4) – 2, 4, 6 or 8 ports
Optical interfaces – 50GbE: SFP56 (LR4/SR4) – 2, 4, 6 or 8 ports
Optical interfaces – 25GbE: SFP28 (SR/LR) – 2, 4, 6 or 8 ports

Datasheet

Overview

The Calnex SNE Ignite is a multi-port, FPGA-based precision network emulator that applies line-rate network impairments on Ethernet links at 400GbE, 100GbE, 50GbE, 25GbE and 10GbE. It is intended for applications that require high accuracy, such as 5G O-RAN O-DU to O-RU (fronthaul) and O-DU to O-CU (midhaul) vendor interoperability testing against O-RAN WG4 and WG5 latency requirements, and for validating high-capacity datacenter interconnect designs before deployment.

O-RAN interoperability

Using packet and protocol impairments, the SNE Ignite can validate minimum and maximum latency performance against the O-RAN T12_max/T34_max specifications in multi-vendor deployments. O-RAN fronthaul links cannot be tested to interoperability specifications unless the emulator's in-line latency is below 8 µs; the SNE Ignite has an intrinsic latency of 6.8 µs. The manufacturer also positions it for O-RAN WG11 cybersecurity and robustness tests and TIFG.E2E end-to-end tests.

Timing and synchronization

An integrated one-step PTP Transparent Clock allows data flows to be separated and impaired independently without affecting synchronization, regardless of packet size. SyncE pass-through preserves physical-layer frequency synchronization between devices, and an external clock input allows synchronization to a remote frequency reference.

User interface and automation

The SNE Ignite uses the graphical, map-based Web UI of the Calnex SNE platform: impairments are dragged onto a network map and linked in the required order. Automation is available through the Local Timeline Option or a RESTful API, and multiple concurrent users are isolated through port reservation.

Key Features

FPGA full line-rate throughput for all impairment and transparency features
Line-rate network emulation at 400GbE, 100GbE, 50GbE, 25GbE and 10GbE
Configurable as two, four, six or eight 100GbE/50GbE/25GbE/10GbE ports, or four 400GbE ports
Software upgradeable to 400GbE interface speed option
Low in-line intrinsic latency of 6.8 μs
Sub 100 ns PDV and jitter to maintain downstream device lock
Nanosecond resolution and repeatability
Delay at line rate of up to 280 ms at 400GbE and up to 42 seconds at reduced bandwidth with Extended Delay Option
One-step PTP (IEEE1588v2) Transparent Clock with G.8273.3 Class B performance
SyncE pass-through mode maintains the SyncE clock link between devices
SyncE and external clock inputs for physical-layer synchronization to a remote frequency reference
Delay, jitter, drop, error, reorder, duplicate, modify or fuzz Ethernet packets
Packet modification with optional L2, L3 and L4 checksum regeneration
Multi-user support with isolation through port reservation
Graphical map-based drag-and-drop Web UI
Automation via Local Timeline Option or RESTful API

Specifications

Ports

Optical interfaces – 400GbE	QSFP-DD (DR4/FR4/LR4/SR4/SR8/LR8/ER8) – 4 ports, 1 bidirectional path optional
Optical interfaces – 100GbE	QSFP28 (LR4/SR4) – 2, 4, 6 or 8 ports optional
Optical interfaces – 50GbE	SFP56 (LR4/SR4) – 2, 4, 6 or 8 ports optional
Optical interfaces – 25GbE	SFP28 (SR/LR) – 2, 4, 6 or 8 ports optional
Optical interfaces – 10GbE	SFP+ (SR/LR) – 2, 4, 6 or 8 ports optional; the web page states up to four pairs of 100 GbE/25 GbE/10 GbE interfaces
Optical adapters	Optical Adapters are provided to convert QSFP28 to SFP28 for 25GbE, and QSFP28 to SFP+ for 10GbE
Forward Error Correction (FEC) – 400GbE PAM4	RS (544,514) KP4
Forward Error Correction (FEC) – 100GbE NRZ	RS (528,514) KR4 FEC can be disabled at 100GbE
Forward Error Correction (FEC) – 50GbE PAM4	RS (544,514) KP4
Forward Error Correction (FEC) – 25GbE NRZ	RS (528,514) KR4 FEC can be disabled at 25GbE
Clock reference	Internal or External (10 MHz)
SyncE pass thru	Clock can be recovered on one port and passed transparently to another
Passthrough pause frames	As per IEEE 802.3x
Passthrough PFC frames	As per 802.1Qbb

Flows

Impairment profiles	16 profiles allowing 8 flows of impaired packets in each direction. Each profile can be configured independently. 25GbE/10GbE
Filtering	MAC Source and Destination Address, Length/Type; VLAN (Priority, VLAN ID & Type), Custom VLAN; Custom VLAN Length, Offset, Mask, Value; MPLS Label, Custom MPLS; Custom MPLS Length, Offset, Mask, Value; IPv4 Source and Destination Address; IPv4 Version No, DiffServ/ToS, Protocol; IPv6 Source and Destination Address; UDP/TCP Source port, Destination port; Custom L3/L4/L5 Length, Offset, Mask, Value User-configurable filters including ranges and wildcards
GTP filter	Targeted GTP control message impairments (e.g. create session request, modify bearer request, etc.); GTP Version (v1, v2), Type, Tunnel EndPoint ID
eCPRI filter	eCPRI Revision, Concatenation Indicator, Message Type, Encapsulation and Custom Targeted eCPRI impairments
RoE filter	Message Type, Flow ID, Length, Ordering Info Targeted RoE (Radio over Ethernet) impairments
PTP filter	Layer 2 Ethernet, Layer 3/4 UDP IPv4
PTP Transparent Clock	1-step Transparent Clock The residence time is measured and added directly into the correction field of the outgoing PTP event message
PTP Transparent Clock performance	G.8273.3 Class B

General

Web browser user interface	Intuitive drag and drop user interface allowing user to draw out their target network on screen, add impairments as required and visualize the network-under-test
Local automation	Local automation Timeline Option enables local automation to a 1s resolution
Remote control	RESTful API enables full remote control
Smart start-up	Automatically launch previous map on boot
Multi-user support	Multiple users, share maps, assign ports to individual users

Impairments

Packet modification – position	Select from byte 0 to byte 127 within the frame With optional Layer 2, Layer 3, Layer 4 checksum regeneration
Packet modification – granularity	1 bit to 4 bytes Any combination of 1 – 4 packet modifier types can be applied in any order to each Profile
Packet modifier types	Overwrite, Invert, Increment, Decrement, Fuzz (PRBS with a user defined seed value)
Packet modification mode	burst (1–10,000), rate (0.00001 to 100%) Continuous or On/Off/Repeat (applies to all packet types) based on time or number of packets
Delay and jitter distributions	Constant Delay; Gaussian, Gamma (internet), Uniform and Step distribution of jitter; Import from an external file Independent delay/jitter distribution can be added to each profile simultaneously; readout of Max, Min Jitter and Max Delay
Jitter range	1 µs to 400 ms step size is 1 ns
Timeline	Automate emulation and schedule changes to emulation settings via the user interface; timeline can be looped for continuous playback
Minimum delay emulation (intrinsic delay)	6.8 µs at 10GbE/25GbE/50GbE/100GbE/400GbE
Minimum delay emulation (intrinsic delay), Jumbo Frame Mode	14 µs at 10GbE; 8.85 µs at 25GbE/50GbE/100GbE; 6.8 µs at 400GbE Jumbo Frame Mode
Line rate delay	40ms at 400GbE, 80 ms at 100GbE, 160 ms at 50GbE, 320 ms at 25GbE, and 750 ms at 10GbE Full line-rate delay; delay resolution 1 ns
Extended delay emulation (line rate) Extended Delay Option	280ms at 400GbE, 320 ms at 100GbE, 640 ms at 50GbE, 1280 ms at 25GbE, and 3200 ms at 10GbE Extended Delay Option; delay resolution 1 ns
Extended delay emulation (reduced bandwidth) Extended Delay Option	up to 22 s (all rates at 1G reduced bandwidth); up to 42 s (all rates at 500M reduced bandwidth) Extended Delay Option
Delay resolution	1 ns
Link flap	Enables the port administrative state to be cycled up/down, down/up, available via the RESTful API
PDV and jitter	Sub 100 ns

Monitor/Analysis

Map statistics	Port and impairment node level packet statistics
Statistics refresh	Statistics can automatically refresh every 1s or are available on demand
Statistics export	Downloadable via the UI in .csv format; downloadable via the API in .json format

Physical Dimensions

RU height	3.7RU in height
Height	Metric: 170 mm, Imperial: 6.7"
Width	Metric: 440 mm, Imperial: 17.32"
Depth	Metric: 550 mm, Imperial: 21.65"
Weight	Metric: 14 kg, Imperial: 30 lb 13.8 oz
Active chassis cooling	Front to rear
Active PSU cooling	Side to rear, self-contained
Total chassis airflow	Maximum 452 CFM

Specifications are taken from the manufacturer's documentation. See the datasheet for complete and current values.

Applications

O-RAN fronthaul (O-DU to O-RU) and midhaul (O-CU to O-DU)
O-RAN WG4 fronthaul delay management tests
O-RAN WG5 midhaul F1 error handling tests
O-RAN WG11 cybersecurity and robustness tests
O-RAN TIFG.E2E fronthaul/midhaul throughput tests
O-RAN 5G Mobile Edge Computing (MEC)
5G core and backhaul
Datacenter interconnect and migration
Cloud migration planning
Telecom and federal applications
Defense applications
Customer proof of concept (PoC)
SLA verification
IPTV and video

Included

Optical adapters to convert QSFP28 to SFP28 for 25GbE and QSFP28 to SFP+ for 10GbE

Options & Accessories

Option	Description
Extended Delay Option	Extended Delay Option — Extends line-rate delay to 280ms at 400GbE, 320 ms at 100GbE, 640 ms at 50GbE, 1280 ms at 25GbE and 3200 ms at 10GbE; up to 42 s at reduced bandwidth
Local Timeline Option	Local Timeline Option — Local automation to a 1s resolution
400GbE interface speed option	400GbE interface speed option — Software upgrade to 400GbE interface speed