NVIDIA Spectrum SN2100 Ethernet Switch

NVIDIA Spectrum SN2100 Ethernet Switch

Brand: NVIDIA | Category: Network Switches

SKU: NVID-MSN2100CB2FO | Part #: MSN2100-CB2FO | MPN: MSN2100-CB2FO

Contact for Pricing — Request a Quote

Request a Quote Contact Us

About the NVIDIA Spectrum SN2100 Ethernet Switch

The NVIDIA Spectrum SN2100 is a 16-port 100GbE Ethernet switch based on the first-generation Spectrum ASIC, offering 3.2Tb/s of aggregate non-blocking switching bandwidth in an ultra-compact half-width 1U form factor. It is the smallest and most accessible switch in the NVIDIA Spectrum SN2000-series, designed for use cases requiring full Spectrum capabilities at a smaller scale. The SN2100's compact chassis supports side-by-side dual-switch installation in a single rack unit, making it highly efficient for high-density rack environments where space is at a premium.

Despite its compact form factor, the SN2100 delivers the full Spectrum feature set including RoCEv2 RDMA support, EVPN/VXLAN, BGP, OSPF, and MLAG when paired with a second unit. It runs NVIDIA Cumulus Linux, enabling network-as-code automation through Ansible, Puppet, and other infrastructure automation frameworks. Each port supports 4x25GbE breakout, making the SN2100 an effective ToR switch for smaller server pods with 25GbE NIC connectivity.

The SN2100 is widely used for out-of-band management networks in large data centers, small cluster deployments, test and development environments, and proof-of-concept labs. It is also selected for edge computing deployments where full-featured L3 switching in a compact, low-power form factor is required. The switch's support for Cumulus Linux makes it a natural choice for organizations standardizing on a single NOS across their entire switching infrastructure, from edge to core.

Ideal for

  • Out-of-band (OOB) management network switching in large enterprise data centers
  • Top-of-rack (ToR) switch for small GPU server pods and development clusters
  • Proof-of-concept and network automation lab switching with full Cumulus Linux support
  • Edge computing node switching in space-constrained micro data centers
  • Small-scale RoCEv2 RDMA fabric for AI inference or small training workloads
  • Dual-switch MLAG deployment in half-width chassis for high-availability ToR connectivity

Technical specifications

ManufacturerNVIDIA
Part NumberMSN2100-CB2FO
Switch ASICNVIDIA Spectrum
Ethernet Speed100GbE per port (native)
Total Ports16 x QSFP28 100GbE ports
Breakout Support16 x 100GbE or 64 x 25GbE (4x25G breakout per port)
Aggregate Bandwidth3.2 Tb/s non-blocking
Forwarding Rate2.38 Bpps
Latency<600 ns port-to-port (cut-through mode)
Buffer ArchitectureShared and dedicated buffers (Spectrum ASIC)
RDMA SupportRoCEv2 with PFC and ECN support
Layer 2 FeaturesVLAN, MSTP/RSTP, LACP, MLAG, VxLAN, EVPN
Layer 3 FeaturesBGP (EVPN), OSPF, ECMP, BFD, PIM, IGMP snooping
Management Interface1 x 10/100/1000 RJ-45 OOB; USB console
Operating SystemNVIDIA Cumulus Linux
TelemetrySNMP, sFlow, NVIDIA NetQ (where supported)
Form FactorHalf-width 1U rack-mount (two units fit side-by-side in 1U)
Dimensions44 x 217 x 430 mm (H x W x D)
Power SupplySingle hot-swap PSU, 350W typical
CoolingHot-swap redundant fans
Operating Temperature0°C to 40°C
StandardsIEEE 802.3bj (100GbE), 802.1Q, 802.3ad, 802.3x, 802.1X

Available from Omnixon Global. Submit an RFQ and our team will confirm configuration and availability for your order.

Technical Specifications

BrandNVIDIA
CategoryNetwork Switches
SKUNVID-MSN2100CB2FO
Part NumberMSN2100-CB2FO
ConditionNew
Switch ASICNVIDIA Spectrum
Ethernet Speed100GbE per port (native)
Total Ports16 x QSFP28 100GbE ports
Breakout Support16 x 100GbE or 64 x 25GbE (4x25G breakout per port)
Aggregate Bandwidth3.2 Tb/s non-blocking
Forwarding Rate2.38 Bpps
Latency<600 ns port-to-port (cut-through mode)
Buffer ArchitectureShared and dedicated buffers (Spectrum ASIC)
RDMA SupportRoCEv2 with PFC and ECN support
Layer 2 FeaturesVLAN, MSTP/RSTP, LACP, MLAG, VxLAN, EVPN
Layer 3 FeaturesBGP (EVPN), OSPF, ECMP, BFD, PIM, IGMP snooping
Management Interface1 x 10/100/1000 RJ-45 OOB; USB console
Operating SystemNVIDIA Cumulus Linux
TelemetrySNMP, sFlow, NVIDIA NetQ (where supported)
Form FactorHalf-width 1U rack-mount (two units fit side-by-side in 1U)
Dimensions44 x 217 x 430 mm (H x W x D)
Power SupplySingle hot-swap PSU, 350W typical
CoolingHot-swap redundant fans
Operating Temperature0°C to 40°C
StandardsIEEE 802.3bj (100GbE), 802.1Q, 802.3ad, 802.3x, 802.1X

Frequently Asked Questions about NVIDIA Spectrum SN2100 Ethernet Switch

What does the NVIDIA Spectrum SN2100 Ethernet Switch do?

The NVIDIA Spectrum SN2100 Ethernet Switch is enterprise networking hardware for data-center, campus, and branch deployments. Common roles include core/distribution switching, server uplinks, top-of-rack fabric, and edge connectivity in mixed Cisco/Juniper/Arista environments.

What are the headline specs of the NVIDIA Spectrum SN2100 Ethernet Switch?

Key specifications for the NVIDIA Spectrum SN2100 Ethernet Switch: new condition; manufacturer NVIDIA; part number MSN2100-CB2FO; switch asic NVIDIA Spectrum; ethernet speed 100GbE per port (native); total ports 16 x QSFP28 100GbE ports; breakout support 16 x 100GbE or 64 x 25GbE (4x25G breakout per port). Manufacturer part number MSN2100-CB2FO. For the full datasheet with electrical, environmental, and compliance details, contact our pre-sales engineering team.

Is the NVIDIA Spectrum SN2100 Ethernet Switch compatible with my infrastructure?

The NVIDIA Spectrum SN2100 Ethernet Switch interoperates with standard switching protocols (LACP, MLAG/VPC, 802.1Q, OSPF/BGP). For specific cross-vendor scenarios — Cisco-to-Arista MLAG peering, fabric integration with Cumulus/SONiC, or migration from legacy Catalyst — our network engineers map a compatibility plan as part of the RFQ.