High-Density 25GE Access

· The switch offers high-density 100G/40G/25G/10G ports and a wire-speed forwarding capacity as high as 4 Tbps. With standard 25G ports, it can provide high-density server access in high-end data centers.

IRF2 (Second Generation Intelligent Resilience Architecture)

· Facing the application requirements of the unified switching architecture of the data center, the series switches support the IRF2 technology, which virtualizes multiple devices into one logical.

· The equipment has strong advantages in scalability, reliability, distributed and availability.

· IRF2 not only can achieve a long-distance intelligent elastic architecture within a rack, across racks, and even across regions.

Abundant Data Center Features

The switch supports abundant data center features, including:

· H3C S6850 switch series supports VXLAN (Virtual Extensible LAN), which provides two major benefits, higher scalability of Layer 2 segmentation and better utilization of available network paths.

· H3C S6850 switch series supports MP-BGP EVPN (Multiprotocol Border Gateway Protocol Ethernet Virtual Private Network) which can run as VXLAN control plane to simplify VXLAN configuration, eliminate traffic flooding and reduce full mesh requirements between VTEPs via the introduction of BGP RR.

· H3C S6850 switch series support Fiber Channel over Ethernet (FCoE), which permits storage, data, and computing services to be transmitted on one network, reducing the costs of network construction and maintenance.

· H3C S6850 switch series support Priority-based Flow Control (PFC), Enhanced Transmission Selection (ETS) and Data Center Bridging eXchange (DCBX). These features ensure low latency and zero packet loss for FC storage, RDMA applications and high-speed computing services.

H3C Distributed Resilient Network Interconnection (DRNI)

· H3C S6850 switch series support DRNI(M-LAG), which enables links of multiple switches to aggregate into one to implement device-level link backup. DRNI is applicable to servers dual-homed to a pair of access devices for node redundancy.

· Streamlined topology: DRNI simplifies the network topology and spanning tree configuration by virtualizing two physical devices into one logical device.

· Independent upgrading: The DR member devices can be upgraded independently one by one to minimize the impact on traffic forwarding.

· High availability: The DR system uses a keepalive link to detect multi-active collision to ensure that only one member device forwards traffic after a DR system splits.

Powerful Visibility

With the rapid development of data center, the scale of the data center expands rapidly; reliability, operation and maintenance become the bottleneck of data center for further expansion. H3C S6850 switch series conform to the trend of automated data operation and maintenance, and support visualization of data center.

· INT (Inband-Telemetry) is a network monitoring technology used to collect data from the device. Compared with the traditional network monitoring technology featuring one query, one reporting, INT requires only one-time configuration for continuous data reporting, thereby reducing the request processing load of the device. INT can collect timestamp information, device ID, port information, and buffer information in real time. INT can be implemented in IP, EVPN, and VXLAN networks.

· Provides a variety of traffic monitoring and analytic tools, including sFlow, NetStream, SPAN/RSPAN/ERSPAN mirroring, and port mirroring to help customers perform precise traffic analysis and gain visibility into network application traffic. With these tools, customers can collect network traffic data to evaluate network health status, create traffic analysis reports, perform traffic engineering, and optimize resource allocation.

· Supports realtime monitoring of buffer and port queues, allowing for visible and dynamic network optimization.

· Supports PTP (Precision Time Protocol) to achieve highly precise clock synchronization.

RoCE (RDMA over Converged Ethernet)

· Remote Direct Memory Access (RDMA) directly transmits the user application data to the storage space of the servers, and uses the network to fast transmit the data from the local system to the storage of the remote system. RDMA eliminates multiple data copying and context switching operations during the transmission process, and reduces the CPU load.

· RoCE supports RDMA on standard Ethernet infrastructures. H3C S6850 switch support RoCE and can be used to build a lossless Ethernet network to ensure zero packet loss.

· RoCE include the following key features，include PFC(Priority based Flow Control), ECN(Explicit Congestion Notification), DCBX(Data Center Bridging Capability Exchange Protocol), ETS(Enhanced Transmission Selection).

Flexible programmability

· The switch uses industry-leading programmable switching chips that allow users to define the forwarding logic as needed.

· Users can develop new features that meet the evolving trend of their networks through simple software updates.

Powerful SDN capacity

· H3C S6850 switch series adopt the next-generation chip with more flexible Openflow FlowTable, more resources and accurate ACL matching, which greatly improves the software-defined network (SDN) capabilities and meet the demand of data center SDN network.

· H3C S6850 switch series can interconnect with H3C SeerEngine-DC Controller through standard protocols such as OVSDB, Netconf and SNMP to implement network automatic deployment and configuration.

Comprehensive security control policies

· H3C S6850 series switch supports AAA, RADIUS and user account based authentication, IP, MAC, VLAN, port-based user identification, dynamic and static binding; when working with the H3C iMC platform, it can conduct real time management, instant diagnosis and crackdown on illicit network behavior.

· H3C S6850 series switch supports enhanced ACL control logic, which enables an enormous amount of inbound and outbound ACL, and delegate VLAN based ACL. This simplifies user deployment process and avoids ACL resource wastage. S6850 series switch can also take advantage of Unicast Reverse Path Forwarding (Unicast RFP). When the device receives a packet, it will perform the reverse check to verify the source address from which the packets are supposedly originated, and will drop the packet if such path doesn’t exist. This can effectively prevent the source address spoofing in the network.

Multiple reliability protection

· The S6850 series switch provides multiple reliability protection at both switch and link levels. With over current, overvoltage, and overheat protection, all models have a redundant pluggable power module, which enables flexible configuration of AC or DC power modules based on actual needs. The entire switch supports fault detection and alarm for power supply and fan, allowing fan speed to change to suit different ambient temperatures.

· The switch supports diverse link redundancy technologies such as H3C proprietary RRPP, VRRPE, and Smart Link. These technologies ensure quick network convergence even when large amount of traffic of multiple services runs on the network.

·

Flexible choice of airflow

· To cope with data center cooling aisle design, the H3C S6850 series switch comes with flexible airflow design, which features bi-cooling aisles in the front and back. Users may also choose the direction of airflow (from front to back or vice versa) by selecting a different fan tray.

Excellent manageability

The switch improves system management through the following ways:

· Provides multiple management interfaces, including the serial console port, mini USB console port, USB port, two out-of-band management ports, and two SFP ports. The SFP ports can be used as in-band management port through which encapsulated sampling packets are sent to the controller or other management devices for deep analysis.

· Supports multiple access methods, including SNMPv1/v2c/v3, Telnet, SSH 2.0, SSL, and FTP.

· Supports standard NETCONF APIs that allow users to configure and manage the switch, enhancing the compatibility with third-party applications.

Hardware Specification

Software Specification

Performance and scalability