The HPE StoreFabric SN2100M is a high-density, half-width 100 Gigabit Ethernet switch designed for enterprise storage fabric and data center networking. It provides 16 QSFP28 ports supporting 100GbE connectivity with a total switching capacity of 1.6 Tbps and forwarding rate of 2.38 Bpps, enabling non-blocking, wire-speed performance across all ports simultaneously. The switch supports flexible port configurations through QSFP28 breakout cables, allowing up to 64 sub-ports in x4 mode or 32 sub-ports in x2 mode for mixed-speed deployments. Built on Mellanox Spectrum ASIC architecture, it delivers ultra-low latency with true cut-through forwarding, zero packet loss performance, and native RoCEv2 RDMA support for converged storage and compute environments. Advanced Layer 2 and Layer 3 features include VLAN trunking, LACP, STP/RSTP, IPv4/v6 routing with OSPF and BGP, VRF, ECMP, and PTP boundary clock functionality. The SN2100M is optimized for power efficiency with less than 6 watts per port, reducing operational costs while maintaining enterprise-grade reliability. Ideal for primary and secondary storage deployments, hyperconverged infrastructures, and high-performance computing environments requiring deterministic, low-latency fabric connectivity.
1.6Tbps switching capacity with non-blocking architecture.
16x 100GbE QSFP28 ports with x4/x2 breakout support.
Native RoCEv2 RDMA for converged storage networking.
Ultra-low latency with true cut-through forwarding.
Less than 6 watts per port for power efficiency.
Half-width 1RU form factor for dense deployments.
Expedited and overnight shipping options are available at checkout, priced by destination and service.
Purchase orders accepted from qualified institutions.
Refunds within 14 days and replacements within 30 days of delivery with an RMA (25% restocking fee on customer-initiated returns), and every part carries our Standard 30-Day Warranty - see our return policy or contact US IT Hardware.
Featuring
Manufacturer part #
P11673-001
Brand
HPE
Condition
New
Cooling
Front-to-back or back-to-front airflow; optimized for low heat generation
Model Name
SN2100M
Device Type
16-Port Network Switch
Product Line
StoreFabric M-series
Product Type
100 Gigabit Ethernet Switch
RoCEv2 Support
Native Layer 2 and Layer 3 RDMA over Converged Ethernet v2
The SN2100M integrates seamlessly into HPE StoreFabric M-series environments and supports open networking with Cumulus Linux or proprietary ONYX OS. It is compatible with standard 100GbE QSFP28 optics, breakout cables for mixed-speed deployments, and enterprise server NICs. Management is via standard RJ45 Ethernet or RS232 serial console. Power and cooling are configurable for various data center environments.
HPE StoreFabric
Model: SN2100M
Form factor: Half-width 1RU
Generation: M-series
Can be deployed side-by-side with other SN2100M units for increased port density in 1RU space.
HPE StoreFabric
Model: SN2700M
Form factor: Full-width 1RU
Generation: M-series
Compatible in multi-tier fabric architectures; SN2100M typically used as ToR, SN2700M as spine.
HPE StoreFabric
Model: SN2010M
Form factor: Half-width 1RU
Generation: M-series
Lower-speed alternative with 18x 25GbE SFP28 + 4x 100GbE QSFP28 ports for mixed-speed deployments.
Best for
Storage Fabric
Primary and secondary storage array interconnect with RoCEv2 RDMA for NVMe over Fabrics and iSER protocols.
Data Center Spine
Details
Shipping, warranty & returns
Customer reviews
No reviews yet — be the first to review this part.
Free ground shipping over $100
Secure SSL checkout. Visa, Mastercard, Amex and Discover accepted. ACH / wire available for businesses.
Buy with confidence
New, certified-refreshed and OEM stock with guaranteed compatibility, backed by warranty on every part.
PFC (Priority Flow Control), DCBX, ECN, zero packet loss for NVMe TCP and RoCEv2
Network & Communication
Latency
Ultra-low cut-through forwarding
Forwarding Rate
2.38 Bpps (billion packets per second)
Packet Buffering
16 MB fully-shared
Switching Capacity
1.6 Tbps full-duplex
Power Consumption
Power Supply
Redundant hot-swappable AC or DC power modules (configuration-dependent)
Power Consumption
Less than 6 watts per port; specific total wattage depends on configuration and port utilization
Physical Characteristics
Form Factor
Half-width 1RU (can be deployed side-by-side with another SN2100M in 1RU)
Airflow Direction
Configurable: Power-to-Connector (P2C) or Connector-to-Power (C2P)
System Status Indicators
System, Fan, Power Supply, Unit Identifier LEDs
Environmental
Humidity
5% to 95% non-condensing
Storage Temperature
-40°C to 70°C (-40°F to 158°F)
Operating Temperature
0°C to 40°C (32°F to 104°F)
Interface/Ports
Port Configuration
16x 100GbE QSFP28 ports
Management Interface
1x RJ45 Ethernet (1Gbps) front, 1x RS232 serial console (115200 baud), 1x USB 2.0 Mini
Port Breakout Support
x4 mode (64 sub-ports at 25GbE) or x2 mode (32 sub-ports at 50GbE)
High-speed spine/core layer switch for large-scale data center networks requiring 100GbE uplink capacity.
HPC Networking
High-performance computing cluster interconnect with low-latency, zero-loss forwarding for scientific computing.
Hyperconverged Infrastructure
Converged storage and compute fabric for HCI platforms combining storage, compute, and networking.
Database Acceleration
Low-latency interconnect for in-memory databases and real-time analytics requiring deterministic performance.
Cloud Data Center
Tenant isolation and east-west traffic optimization in multi-tenant cloud environments.
Frequently asked questions
What is the switching capacity and forwarding rate of the SN2100M?
The HPE StoreFabric SN2100M delivers 1.6 Tbps full-duplex switching capacity with a forwarding rate of 2.38 Bpps (billion packets per second). This non-blocking architecture ensures all 16 ports can operate at full 100GbE line rate simultaneously without congestion, making it ideal for demanding storage fabric and data center spine applications requiring deterministic performance.
Does the SN2100M support port breakout for mixed-speed deployments?
Yes, all 16 QSFP28 ports support split x4 or x2 mode using breakout cables. In x4 mode, you can achieve up to 64 sub-ports at 25GbE each; in x2 mode, up to 32 sub-ports at 50GbE each. This flexibility allows you to mix 100GbE uplinks with lower-speed access ports in the same switch for cost-optimized, multi-tier fabric designs.
What RoCEv2 and RDMA capabilities does the SN2100M provide?
The SN2100M offers native RoCEv2 (RDMA over Converged Ethernet v2) support at both Layer 2 and Layer 3, enabling zero-copy, low-latency remote memory access for NVMe over Fabrics, iSER, and other RDMA-accelerated protocols. It includes full PFC (Priority Flow Control), DCBX, and ECN support for lossless Ethernet, critical for storage and HPC workloads.
What Layer 3 routing protocols are supported?
The SN2100M supports comprehensive IPv4/v6 routing including OSPFv2, OSPFv3, BGPv4/v6, VRF with route leaking, ECMP with resilient hashing, BFD, policy-based routing, GRE tunneling, and PTP boundary clock. These features enable it to function as a spine switch in large-scale, multi-tier data center networks with advanced traffic engineering.
How does the SN2100M compare to the SN2700M in terms of port count and capacity?
The SN2100M provides 16x 100GbE QSFP28 ports (1.6Tbps capacity), while the SN2700M offers 32x 100GbE QSFP28 ports (6.4Tbps capacity). The SN2100M is a half-width switch ideal for top-of-rack and smaller fabric deployments, whereas the SN2700M is a full-width switch for larger spine/core roles requiring higher port density and throughput.
What is the power consumption and thermal profile of the SN2100M?
The SN2100M is designed for ultra-low power consumption at less than 6 watts per port, significantly reducing operational expenses and heat generation compared to competing switches. This efficiency, combined with its compact half-width form factor, allows for higher density deployments in power-constrained data center environments while maintaining enterprise reliability.
What packet buffering and latency characteristics does the SN2100M offer?
The SN2100M includes 16 MB of fully-shared packet buffering and delivers true cut-through forwarding with ultra-low latency, enabling zero-packet-loss performance for NVMe TCP and RoCEv2 RDMA workloads. This combination of ample buffer and minimal latency is critical for storage fabric and HPC applications where predictable, deterministic performance is essential.
Cisco MerakiMS125-48LP-HW
Allied TelesisAT-8624T/2M
Allied TelesisAT-9000/24
Dell469-3414
Allied TelesisAT-GS910/24-50
Allied TelesisAT-FS717FC/SC-10
Allied TelesisAT-9424T/SP