|
Advanced CLOS+ multi-level multi-plane switching architecture
* Advanced CLOS/CLOS+ multi-level multi-plane orthogonal switching architecture with no backplane/no midplane design to provide continuous bandwidth upgrade capability.
* Supports high-density 40GE and 100GE Ethernet ports to fully meet the needs of data center applications and future development.
* Independent switching network board, control engine and switching network board hardware are independent of each other, maximizing equipment reliability and guaranteeing continuous upgrade of subsequent product bandwidth.
* Dynamic variable cell fragmentation is strictly non-blocking switching, improving overall forwarding performance.
Full range of virtualization capabilities
* IRF2 (Intelligent Resilient Framework2, second generation intelligent elastic architecture)
The S12500 series supports IRF2 technology and virtualizes up to four high-end devices into one logical device. It has strong advantages in reliability, distribution and manageability. It supports ultra-long-distance IRF2 across transmission wavelength division equipment to meet different locations. Data center dual-active backup solution design requirements.
Reliability: Through the patented hot backup technology, redundant backup and uninterrupted data forwarding of all information on the control plane and data plane are realized in the entire virtual architecture, which greatly enhances the reliability and high performance of the virtual architecture while eliminating A single point of failure, avoiding business interruptions;
Distribution: Through the distributed cross-device link aggregation technology, load balancing and mutual backup of multiple uplinks are implemented, thereby improving the redundancy of the entire network architecture and the utilization of link resources;
Manageability: The entire flexible architecture shares an IP management, simplifying network device management, simplifying network topology management, improving operational efficiency, and reducing maintenance costs.
* MDC (Multitenant Devices Context, multi-tenant device environment)
The S12500 series can achieve 1:N virtualization capability through MDC technology, that is, a physical switch is virtualized into N logical switches, and up to 16 logical switches can be virtualized to meet the needs of multi-client shared core switches; The ports of the board are divided into different MDCs, which can fully utilize the capabilities of the core switches and reduce the investment cost of users. The MDC technology is used to achieve security isolation of services.
* IRF3.1 (Intelligent Resilient Framework 3.1, the third generation of intelligent elastic architecture upgrade)
The S12500 series supports heterogeneous virtualization in the vertical dimension. The core and access devices form a vertical logical virtual device through IRF3.1 technology, which is equivalent to adding a box device as a remote interface board to the main device system. In order to achieve the expansion of I / O port capabilities and centralized control management purposes. IRF3.1 technology can simplify management, greatly reduce network management nodes; simplify wiring, layer 2 becomes a layer, saves horizontal connection cables; finally realizes data forwarding plane virtualization, which simplifies service deployment and automatic orchestration.
* EVB (Edge Virtual Bridging Edge Virtual Bridge)
The S12500 series supports Edge Virtual Bridging (EVB). The virtual Ethernet port Aggregator (VEPA) mode is used to upload data traffic generated by a virtual machine VM to a physical switch connected to the server for processing. Not only does the traffic forwarding between virtual machines be realized, but also the connection and management boundary between the virtual machine and the network is solved.
Application requirements for cloud computing data centers
* TRILL/SPB: As the size of servers and switches increases, data center networks are increasingly oriented toward flattened network architectures for maintenance management, which requires the construction of a large Layer 2 network; the S12500 series supports TRILL or SPB technology to build a large second-tier network in the data center. Data Center's Translatent Interconnection of Lots of Links (TRB) and SPB (Shortest Path Bridging) protocol combine the simplicity, flexibility and stability of the Layer 2 layer with the stability and scalability of the Layer 2 layer. High-performance organic integration provides higher density ports and a flatter network architecture to meet the access needs of large-scale servers in the data center.
* The S12500 series supports VXLAN (Virtual eXtensible LAN) technology. VXLAN is a Layer 2 VPN technology based on IP network and in "MAC in UDP" encapsulation. VXLAN can provide Layer 2 interconnection for distributed physical sites based on existing service providers or enterprise IP networks, and can provide service isolation for different tenants.
The S12500 series supports EVPN (Ethernet Virtual Private Network). EVPN is a Layer 2 VPN technology. The control plane uses MP-BGP to advertise EVPN routing information. The data plane supports VXLAN encapsulation to forward packets.
* S12500 series products support EVI (Ethernet Virtual Interconnection) technology. EVI is an advanced "MAC in IP" technology. EVI solution deployment is very simple, based on existing IP networks, for distributed physics. The site provides flexible Layer 2 connectivity.
* The S12500 series supports FC and FCoE (FC over Ethernet) technology. The FCoE technology is mainly used to solve the heterogeneity of the LAN network and the FC storage network in the cloud computing data center. Through the deployment of FCoE and CEE technologies, Realize the convergence of the data center front-end network and the back-end network architecture to solve the technical problems of data, computing and storage splitting, thus greatly reducing the procurement and expansion costs of the data center.
* The S12500 series supports large-capacity ARP/ND, MAC, and ACL entries, which can be adapted to the flat networking requirements of large data center networks.
Innovative multi-engine control design
* With innovative hardware design, the system provides powerful control and 50ms high reliability through independent control engine, detection engine and maintenance engine.
* Independent control engine, providing a powerful master CPU system, easily handling various protocol messages and control messages, and supporting fine control of protocol packets, providing the system with a comprehensive ability to resist protocol packet attacks;
* Independent detection engine provides high reliability and high performance FFDR (Fast Fault Detection and Restoration) system for fast fault detection such as BFD and OAM, and linkage with the control plane protocol to support fast Protection switching and fast convergence can achieve 50ms fault detection and ensure uninterrupted service;
* Independent maintenance engine, intelligent EMS (Embedded Maintenance Subsystem) CPU system, which supports power supply intelligent management, can support the board to power on and off sequentially (reduce the power supply of the board at the same time) Impact, improve equipment life, reduce electromagnetic radiation, reduce system power consumption), equipment online status check.
* The first cloud master with integrated X86 architecture, support 256G SSD hard drive, Intel Xeon E5 8 core processor, 64G memory; support SDN controller / Director / third-party software, realize double plane management control of data center Overlay & Underlay Support for installing Linux/Windows systems, support for installing virtual machine/NFV functions, and supporting docking with Openstack/third-party cloud platforms.
Data center level reliability assurance
* The S12500 series provides a dedicated FFDR system for fast fault detection such as BFD and OAM, and is linked with the protocol of the control plane to support fast protection switching and fast convergence.
* Supports BFD for VRRP/BGP/IS-IS/RIP/OSPF/RSVP/static routing.
* Supports NSR/GR for OSFP/BGP/IS-IS/RSVP.
* Control engine and switching network board hardware are independent of each other, realizing physical separation of control plane and forwarding plane, control engine 1+1 redundancy; switching network board N+M redundancy; fan frame 1+1 redundancy; power module N+ M redundancy; maximizes system fault isolation and reliability.
Multi-level security
* Multi-level protection and security of the control plane. The control plane policy feature of the S12500 manages the packet flow from the data plane (DP) to the control plane (CP) by configuring QoS filtering and rate limiting to protect the S12500 switch from DDoS attacks. It can identify and protect important packets and discard illegal packets to ensure that the control plane maintains normal forwarding and protocol status in case of attacks or large traffic.
* Supports a large number of ACL rules and meets full line rate forwarding; it can perform fine security access control on various L2/IPv4/IPv6/MPLS packets and their field combinations.
* Supports hardware-level encryption Macsec technology (802.1ae), which is different from the traditional end-to-end basic application layer protection soft encryption technology. Macsec protects the management of bridged networks and other data by protecting the data source's cryptographic technology to protect information integrity and Provide reprotection and confidentiality services. By confirming the frame sent by the station, the protection layer 2 protocol can be attacked at the root.
Distributed caching mechanism and refined QoS
* Innovative adoption of "distributed ingress caching" technology in the face of next-generation data center bursts. Each port can accurately perform accurate bandwidth allocation and traffic shaping for all traffic flows to the port. The precise scheduling of the forwarding plane ensures distributed buffering in the Ingress direction, effectively sharing and utilizing the caches distributed on each line card. Space, providing better caching.
* The network application model has been converted from C/S to B/S model. The change of application mode has led to the increase of network burst traffic. The large cache mechanism has become an urgent need of network equipment. The S12500 supports 200ms burst traffic per 10Gb port, combined with a distributed ingress cache mechanism to meet the high burst traffic requirements of large data centers.
* Single chip supports 4GB cache, line card supports up to 32GB cache.
* Maximum 96K hardware queues per board, support for refined QoS and traffic management, and can accurately assign different priorities and queues to different users and different service flows according to requirements, ensuring different bandwidth, service delay and jitter performance.
Comprehensive maintenance inspection mechanism
* Online status detection mechanism, through the dedicated maintenance engine, can realize the detection of the switching network board, backplane communication channel, service communication channel, key chip, memory, etc. of the device. Once the relevant module fails, it is reported to the system through the EMS.
* The board isolation function can isolate the specified board from the forwarding plane and no longer participate in the forwarding plane forwarding. However, the isolated board is still in the control plane and can be managed. You can perform real-time diagnosis and CPLD upgrade on the board without affecting the services of the entire system.
* Supports Ethernet OAM, providing multiple device-level and network-level fault detection methods.
Open application architecture
* The S12500 series products are based on the OAA (Open Application Architecture) concept and innovatively launched an open business platform.
* Support network security multi-service card module, can integrate firewall module, IPS module, ACG application control gateway module, LB load balancing module. A network security integrated solution through product integration.
Green design
* The S12500 series supports the intelligent management of the power supply through the intelligent EMS engine system. It can support the sequential power-on of the boards (reducing the power supply shock caused by the simultaneous power-on of the boards, improving the life of the equipment and reducing the electromagnetic radiation). Power off the board and isolate faulty/idle boards to reduce system power consumption.
* S12500 series products use high-efficiency PWM speed control fans to support stepless speed regulation. The system can automatically collect the temperature of the board and calculate the fan speed curve based on the actual situation of the device. The system supports fan status monitoring (speed monitoring, fault alarm, etc.). It can automatically adjust the speed according to the ambient temperature and board configuration to reduce equipment power consumption and running noise, effectively reduce environmental noise and extend fan life.
* The S12500 series supports automatic detection of internal ports. When an interface board is not configured in a slot or when a port is not connected to a cable, the system can automatically shut down the corresponding internal port, saving power consumption.
* The minimum power consumption of the 10G port is less than 4.5W, the minimum power consumption of the 40G port is less than 8W, and the minimum power consumption of the 100G port is less than 23W.
* S12500 series products meet the EU RoHS standards for environmental protection and safety of materials.
* S12500 series products adopt front and rear ventilation design, which can meet the requirements of efficient cooling of data center equipment room.
|
