The Cisco Nexus 9300-EX and 9300-FX platforms, which are based on Cisco Cloud Scale technology, are the following iteration of fixed Cisco Nexus Network Switch. The new platforms offer more endpoints, more cost-effective cloud-scale installations, and cloud services with wire-rate security and telemetry.
The systems are constructed using cutting-edge system architecture created to deliver great performance and cater to the changing requirements of highly scalable data centers and expanding businesses.
The following generation of stationary N9K-C93108-EX-B24C – Cisco Nexus Network Switch is the Cisco Nexus® 9300-EX platform. The new platform, which is based on Cisco’s Cloud Scale ASIC, provides cloud services with wire-rate security and telemetry, cost-effective cloud-scale deployments, and an enhanced number of endpoints. The platform’s current system architecture is engineered to deliver excellent performance in order to satisfy the changing requirements of highly scalable data centers and expanding businesses.
In order to upgrade existing data centers from 100 Mbps, 1 Gbps, and 10 Gbps speeds to 25 Gbps at the server and from 10 and 40 Gbps speeds to 50 and 100 Gbps at the aggregation layer seamlessly, a range of interface options are provided by Cisco Nexus Network Switch.
With its massive Layer 2 and Layer 3 scalability, performance, and big buffers, the platform protects clients’ investment while adapting to the changing demands of increasingly virtualized, automated cloud environments in data centers. In order to support Cisco® Data Center Interconnect (DCI) features and provide flexible workload mobility and LAN and SAN convergence, the platform hardware is supported by Cisco.
Characteristics and Advantages
The following characteristics and advantages are offered by the N9K-C93108-EX-B24C – Cisco Nexus Network Switch:
- Support for Cisco ACITM, an industry-leading software defined networking solution.
- Support for VXLAN EVPN fabrics built on standards, including hierarchical multi-site support (refer to VXLAN Network with MP-BGP EVPN Control Plane for more information).
- Three-tier BGP topologies that enable web-scale, non-blocking IPv6 network fabrics.
- Segment routing enables the network to engineer traffic without using Resource Reservation Protocol (RSVP) Traffic Engineering and forward (MPLS) Multiprotocol Label Switching packets (TE). It offers a control-plane substitute for more virtualization and network scalability.
- Full Layer 3 (v4/v6) unicast and multicast protocol support, including BGP, Open Shortest Path First (OSPF), Enhanced Interior Gateway Routing Protocol (EIGRP), Routing Information Protocol Version 2 (RIPv2), Protocol Independent Multicast Sparse Mode (PIM-SM), Source-Specific Multicast (SSM), and Multicast Source Discovery Protocol (MSDP).
- Automation from day one with Power on Auto Provisioning, which significantly cuts down on provisioning time.
- Market-leading integrations for SALT, Ansible, Chef, and other top DevOps configuration management tools.
- RESTCONF/NETCONF provides extensive Native YANG and industry-standard OpenConfig model compatibility.
- All switch CLI functions have widespread APIs (JSON-based RPC via HTTP/HTTPs).
Flexibility and High scalability:
- Flexible forwarding tables on EX models offer up to 1 million shared entries.
- It is possible to define Access Control List (ACL) templates specifically thanks to the flexible use of TCAM space.
Smart Buffer Management
- In the event of link congestion, the platform delivers Cisco’s cutting-edge intelligent buffer management, which has the ability to discern between mouse and elephant flows and apply various queue management strategies to them based on their network forwarding needs.
- Fair Dropping (AFD) using Elephant trap approximation (ETRAP). ETRAP is used by AFD to discriminate between long-lived elephant flows and short-lived mouse flows. In order to ensure that mice flows receive their fair share of bandwidth lacking starved by bandwidth-hungry elephant flows, AFD excludes mice flows from the dropping algorithm. In order to ensure that they receive their fair share of bandwidth, AFD also monitors elephant flows and subjects them to the AFD algorithm in the egress queue.
- ETRAP calculates the incoming flows’ byte counts and compares them to the user-defined ETRAP threshold. A flow transforms into an elephant flow once it passes the threshold.
- With the help of Dynamic Packet Prioritization (DPP), it is possible to divide mouse and elephant flows into two separate queues and allot buffer space to each of them separately. Mice flows, which are sensitive to latency and congestion, can take priority queue and prevent reordering, which enables elephant flows to use the entire link’s bandwidth.
High availability of both hardware and software:
- Layer 2 multipathing is made possible by Virtual Port-Channel (vPC) technology, which does away with Spanning Tree Protocol. Additionally, it permits Layer 2 logical topologies that are more straightforward and completely exploited bisectional bandwidth without altering the current management and deployment methods.
- The utilization of Layer 3 fat-tree architectures is made possible by the 64-way Equal-Cost MultiPath (ECMP) routing. By preventing network bottlenecks, boosting resilience, and adding capacity with minimal network interruption, this feature aids enterprises.
- Hot and cold patching are features of advanced rebooting.
- The Power-Supply Units (PSUs) and fans used by the switches provide N+1 redundancy.
Cisco Titration Analytics platform support
Without adding any latency to the packets or degrading switch performance, the platform hardware can gather thorough Cisco Titration AnalyticsTM telemetry data at line rate across all of the ports. By default, this telemetry data is exported from the switch’s application-specific integrated circuit every 100 milliseconds (ASIC). Three different types of data make up this information:
- Information about flows: This data includes details on endpoints, protocols, ports, the beginning and finish times of flows, their durations, etc.
- Interpacket Variation: The flow’s interpacket variations are recorded via this data. Time to Live (TTL) variance, IP and TCP flags, payload length, etc. are a few examples.
- Context Details: Information about the context is derived from sources other than the packet header. Examples include changes in buffer usage, packet losses within a flow, associations with tunnel endpoints, etc.
Cisco Network Assurance Engine (NAE)
Cisco NAE uses the power of mathematical models to reason on behalf of the operator at the policy, configuration, and dynamic state level. It continuously checks to see if the network infrastructure is performing in accordance with policy intent. NAE can precisely identify network problems, pinpoint which application or portion of the network is affected, determine the main cause of the issue, and make recommendations for how to resolve it.
Without using any packet data, its continuous verification approach switches Day 2 Operations from a reactive to a proactive posture. By anticipating the effects of changes, minimizing network-related IT issues, and cutting the mean time to fix by up to 66 percent, NAE helps prevent downtime. Network security and segmentation compliance are also supported by NAE.