HPE 100GbE 16QSFP28 Power to Connector Airflow Half Width Switch SN2100M with NVIDIA Cumulus

Lossless Low-latency Connectivity at 100 GbE in a Half-width Form Factor

HPE Storage Switch M-series SN2100M provides predictability within a storage network and consistent throughput regardless of the packet size being transferred, the mixture of ports which are sending data, and even within mixed speed environments.

Provides wire-rate performance with zero packet loss across frame sizes, avoiding any negative impact on applications that could occur with frame loss as unexpected packet loss is unacceptable in modern data centers, especially within a storage network.

Capable of forwarding complete capacity wire-rate performance with zero packet loss across ports concurrently at 100 GbE speeds while transferring data across both Layer 2 and Layer 3 networks.

Can be deployed to support 1/10 GbE ports and is designed to be able to evolve over time to support 25, 50, and 100 GbE speeds. This helps future-proof your network architecture and allows for implementing significant speed changes to the architecture over time.

Enhanced for Storage and Hyperconverged Environments

HPE Storage Switch M-series SN2100M provides high-port density in a single rack unit, allowing for higher capacity and efficiency, simplifying scale-out environments and saving on total cost of ownership (TCO). Unique breakout cables fan out individual switch ports to multiple device ports.

With unique form factors and port counts, these Ethernet switches allow two HPE Storage SN2100M units to be deployed side-by-side allowing for increased density, making it the ideal top-of-rack (ToR) switch.

Designed to use less electric power than competing switches, with less than 95W, HPE Storage Switch M-series SN2100M provides one of the industry’s lowest power draws, producing less heat than competing products and allowing reduced OpEx cost.

Unleashes Flash Storage Performance

HPE Storage Switch M-series SN2100M provides ultra-low latency of under 300 ns port-to-port. This is advantageous for flash storage, which moved latency from storage access to the network, as well as for the burst nature of today’s software-defined and cloud data centers traffic.

Provides enough internal bandwidth that there is sufficient capacity to carry ports at 100 Gbps bandwidth concurrently. This allows the switches to avoid head-of-line blocking which can reduce overall performance and throughput.

Introduces dynamic shared buffers and wire speed performance with no packet loss for predictable data delivery.