Qfx3000 qfabric switch


Published on

Published in: Technology, Business
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Qfx3000 qfabric switch

  1. 1. DATASHEET QFX3000 QFABriC SwiTCH Product Description Product Overview Designed to meet the demands of the modern data center, Juniper Networks® QFX3000 QFabric™ Switch delivers a scalable, high-performance, non-blocking fabric which enablesNext-generation data centers demand traditional Layer 2 and Layer 3 frameworks along with virtualization and convergence. Theexponential scale, performance, QFX3000 QFabric Switch is the embodiment of the QFabric architecture, allowing networkflexibility, and control—along with designers to replace traditional tiered networks with a single-tier fabric that connectsexponential reductions in total cost of compute, storage, network, and services resources as extensions of a low latency network.ownership (TCO). The Juniper NetworksQFabric architecture is a revolutionary By treating computing, storage, services, and network resources as fully fungible poolsapproach to data center networking that can be dynamically and rapidly partitioned without the infrastructure or thethat enables a quantum leap in each of applications knowing details about each other, QFabric technology enables performancethese areas. of data centers to improve at a much faster rate than is possible by making performance improvements to the infrastructure elements alone.The Juniper Networks QFX3000QFabric Switch is the foundation ofthe QFabric architecture—a purpose-built solution that allows the creationof high-performance, cost-effective,dynamic, and easy to manage cloud-ready data centers over a wide rangeof scales using standard computing,storage, and services elements. QFabric Director QFabric Interconnect QFabric Node Figure 1: QFX3000 QFabric Switch topological view 1
  2. 2. The QFX3000 QFabric Switch consists of three separate but Architecture and Key Componentsinterdependent edge, interconnect, and control devices— Juniper with the Node, interconnect, and Director, the QFX3000Networks QFX3500 QFabric Node, QFX3008 QFabric interconnect, QFabric Switch operates as a single logical device—a largeand QFX3100 QFabric Director—that represent the internal elements distributed system with a separate data and control plane. Theseof a traditional switch. independent devices are all managed from a single point, greatly• QFabric Node: in the QFabric architecture, the line cards that simplifying the provisioning, monitoring, and troubleshooting of typically reside within a modular chassis switch become high- the network fabric. density, fixed-configuration, 1 U edge devices that provide The QFabric architecture, as implemented with the QFX3000 access into and out of the fabric. The Node, which can also QFabric Switch, is composed of three distinct planes—the data operate as an independent top-of-rack 10GbE switch, provides plane, the control plane, and the management plane. compute, storage, services, and network access for the QFabric • Data plane: The Node and interconnect devices create the architecture. distributed data plane for the QFabric architecture. All data• QFabric Interconnect: The interconnect represents the typical traffic to and from servers and storage is carried over the data backplane of a modular switch, connecting all Node edge plane. Connectivity between Node and interconnect devices is devices in a full mesh topology. This topology provides the data provided via two or four 40 Gbps link aggregation group (LAG) plane connectivity between all Nodes, with the interconnect interfaces. acting as the high-performance backplane. • Control plane: Separating the data plane from the control plane• QFabric Director: The routing Engines embedded within a has always been a key design principle for Juniper to ensure modular switch are externalized in the QFabric architecture via high reliability. This principle is incorporated in the QFabric the Director, which provides control and management services for architecture, where data and control traffic are carried over the fabric. Deployed in clusters to provide redundancy, Directors two separate networks. The distributed nature of the control provide a single management interface to manage the scalable plane is the key to the QFabric architecture’s scalability and data plane provided by the Node and interconnect devices. simplicity, while providing tremendous reliability by eliminatingThe QXF3000 QFabric Switch supports the interconnection of up any single point of failure in the system. Control plane servicesto 128 Nodes to create a single fabric capable of supporting 6,000 are provided by the Director, which uses a completely separate10GbE ports. The Node and interconnect devices together create out-of-band control plane network that connects Directors andthe distributed data plane for the QFabric architecture over which Nodes to the interconnect and is used exclusively for carryingall of the data traffic to and from servers and storage is carried. control traffic.One of the greatest advantages of QFabric technology is its • Management plane: The Director also provides all managementmanageability. Unlike traditional deployments with multiple services for the QFabric architecture. The Director directlytouch points for provisioning and troubleshooting, the QFabric communicates with all Node and interconnect devices andarchitecture presents a single management interface for builds a global view of the entire network, which may consistprovisioning, managing, and troubleshooting the data center of thousands of server-facing ports. This provides a singleat scale—up to 128 top-of-rack switches connecting network, point of visibility, control, and management for the entire datacompute, and storage resources. center fabric, and it dramatically reduces the operational costs typically associated with managing a data center network. The Director also interfaces with the network management ecosystem via standards-based protocols such as XML/ NETCONF, SNMP, or command-line interface (CLi). Junoscript automation and Juniper Networks Junos® SDK provide a rich automation framework that enables network customization and tuning as required, ensuring the QFabric architecture deployment fits the existing ecosystem without having to invest in special tools. 2
  3. 3. Data Center Deployments FCoE-FCNew revenue models are focusing on iT data centers, which are Gateway Storage SANevolving to address newer architectures such as virtualization, i/O FCoE FCconvergence, and private, public and hybrid clouds. Next-generationdata centers, designed for high performance and scale, are built FCoE Serverswith small form-factor multi-core blade and rack servers. Multi-core processor technologies and high-density memories, whichallow for greater compute capacity and server densities, are drivingtraffic levels higher, creating the need for a high-speed, low latencystorage- and i/O-converged network that can reduce or eliminateinefficiencies and maximize performance in the data center.High-Performance Layer 2/Layer 3 DeploymentsThe QFX3000 QFabric Switch is designed to provide a low latencyfabric that can scale to 6,000 ports and be deployed in a variety FCof environments. with the advent of server virtualization, the iT FCoE Serversinfrastructure is providing business efficiency by consolidating FCoEmany physical servers into fewer high-performance virtualized Figure 2: QFX3000 QFabric Switch as an FCoE transit switch orservers. However, this introduces new challenges in the data center FCoE-to-FC gatewayby drastically increasing network utilization and requiring faster Cloud-Ready Data Center Networkaccess-layer connectivity. Every QFabric Node in a QFX3000 Cloud architectures—both private and public—require high levelsQFabric Switch adds 48 high-performance, ultra-low latency of scalability, elasticity, and multi-tenancy capabilities. For many(ULL) 10GbE ports, making it possible to support large-scale server other enterprises, operational efficiency is absolutely critical. Thevirtualization deployments with a large media access control QFabric architecture can scale from just a few hundred ports to(MAC) address table with ultra-low latency (<5 microseconds port- thousands of server/storage ports, helping customers build highlyto-port) at L2 and L3 from server node to server node. scalable, high-performance, highly efficient cloud-ready (private,The QFX3000 QFabric Switch offers the following advantages for public, or hybrid) data center infrastructures.high-performance access:• Full featured, standards-based L2 and L3 switching capabilities• wire-speed switching on 48 10GbE ports, with sub-microsecond latency and low power consumption at each Node QFabric Interconnect• Scaling to 6,144 10GbE ports, providing CapEx and OpEx QFabric Director efficiencies at scale• Support for the same Juniper Networks Junos operating system Ultra-low latency that powers other Juniper Networks switches, routers, and (<5 microseconds port to port) security products, as well as the Juniper Networks Junos Space AN 25 0 management platform 24 VL A N VL LA VL QFabric NodeVirtualization and I/O Convergence AN V 15 N 00 10As businesses adopt 10GbE connectivity in the access layer, CapEx Inter-VLAN Communicationreduction can be achieved by consolidating storage and Ethernettraffic on common 10GbE server connections. Deployments of Localvirtualized servers hosting hundreds of virtual machines and high- Switchingperformance servers with converged network adapters, as well asstorage and i/O convergence, will require low latency and lossless Figure 3: QFX3000 QFabric Switch in a cloud-ready environment10GbE technologies at the server access layer to support FibreChannel over Ethernet (FCoE) and Fiber Channel (FC) interfaces.The QFX3000 QFabric Switch offers full featured Data CenterBridging (DCB) and FCoE on every QFabric Node port. in addition,the QFX3000 QFabric Switch can be an FCoE transit switch or anFCoE-to-FC gateway. 3
  4. 4. FeAtuRe BeNeFIt Multi-terabit capacity and scalability The QFabric architecture allows for scale and incremental growth by adding a QFabric Node device when a new server rack is installed. with the QFabric interconnect acting as the network backplane with a capacity of 40 Tbps, the QFabric architecture scales to support up to 6,000 10GbE ports, with the ability to scale beyond 10GbE in the future. The incremental investment required to connect new server and storage capacity is a fraction of the existing tiered architecture, and the return on investment is much faster than existing deployment models. Low latency fabric The QFabric architecture’s any-to-any connectivity supports high-speed server-to-server communication. QFabric technology has interface-to-interface latency on the order of 900 nanoseconds, and less than 5 microseconds across the fabric between any two interfaces. Ultra-low latency provides an order of magnitude improvement in performance over traditional network architectures, making the QFabric architecture ideal for supporting latency sensitive applications, east-west traffic flows, virtualization, cloud computing, and other high-performance data center initiatives. Virtualization and converged i/O The QFabric architecture is optimized for server virtualization, making it easier for enterprises to eliminate architecture “stranded capital” resulting from the underutilization of existing server and storage assets. QFabric architecture also supports converged traffic at 10GbE access port speeds, enabling enterprises to migrate to 10GbE converged access on a single network. QFabric technology eliminates the need to provision separate networks for LAN and storage area network (SAN) trade-offs that three-tiered tree networks must make with respect to simplicity, performance, scalability, and cost while delivering the same scale, performance, and enhanced features as traditional networks. robust, modular, feature-rich software Each release of Junos OS runs consistently across all Juniper Networks routing platforms and feature sets. Junos OS was conceived and implemented as a modular design, with each Junos OS process running in protected memory to guard against system crashes and to ensure that applications do not interfere with each other. Junos OS provides the greatest breadth of features and most stable network operating system in the industry. Operational performance By presenting itself as a single switch running Junos OS, the QFX3000 QFabric Switch greatly simplifies data center management. And by reducing the number of switches in the data center network, the QFabric architecture significantly lowers complexity and operational expenses, as well as power, space, and cooling costs. in addition, to maximize network uptime, Juniper has designed the QFabric architecture as a reliable carrier-class infrastructure with no single point of failure or downtime for reconfiguration and maintenance. interface flexibility The QFX3000 QFabric Switch provides a range of interfaces for next-generation data center access from 1GbE, 10GbE, nx10GE, and 40GbE* for Ethernet. The QFabric architecture also provides FCoE as well as FC interfaces required for converged i/O environments. This interface variety (both optical and copper) in a single-tier network offers distinct advantages in terms of OpEx and CapEx reduction. Combining the functions of previously disparate compute, storage, and services offers greater network simplicity and retains the service building advantages of the converged network. Scaling efficiency/increased rOi The QFabric architecture provides for significant investment protection and enhances the “pay-as-you- grow” model by eliminating the need for network tiers. Adding compute clusters simply requires deploying QFX3500 Nodes to the fabric. This does not require upfront investment and growth can be spread over multiple phases. As new points of delivery (PoDs) are added to the data center, Nodes can be connected and provisioned into the existing QFabric architecture, reducing CapEx and OpEx while providing efficiencies with scale and performance. Green The QFabric architecture is environmentally conscious, allowing enterprises to optimize every facet of the data center network while consuming less power, requiring less cooling, and producing a fraction of the carbon footprint of legacy multitiered data center networks. The individual components of QFabric technology are also designed to meet green standards such as restriction of Hazardous Substances (roHS), registration, Evaluation, Authorisation and restriction of Chemical substances (rEACH), 80 Plus, waste Electrical and Electronic Directive (wEEE Directive), and others.Junos Operating SystemThe QFX3000 QFabric Switch components run the same reliable track, and employs a highly available modular architecture thatand high-performance Junos operating system that is used by prevents isolated failures from bringing down an entire system.Juniper Networks EX Series Ethernet Switches, Juniper routers, These attributes are fundamental to the core value of theand Juniper Networks SrX Series Services Gateways. By utilizing software, enabling Junos OS-powered products to be updateda common operating system, Juniper delivers a consistent simultaneously with the same software release. Features areimplementation and operation of control plane features across fully regression tested, making each new release a superset ofproducts. To maintain that consistency, Junos OS adheres to a the previous version. Customers can deploy the software withhighly disciplined development process, follows a single release confidence that existing capabilities will be maintained and operate in the same way.* roadmap 4
  5. 5. QFabric Architecture Management Capabilities • QFabric technology is supported by Junos Space, anThe following system management options are available for the open, programmable application platform for hosting aQFabric architecture and the QFX3000 switch: comprehensive suite of network operational application tools that provide a smart, simple, and open approach for automating• The standard Junos OS CLi and SNMP module offers the same the deployment and operation of a Juniper infrastructure. granular management capabilities and scripting parameters • The QFabric architecture also supports Junoscript and found in any router or switch powered by Junos OS. Junoscript automation. The Junoscript APi facilitates• Performance, configuration, and fault data for the QFabric application and script development to develop custom architecture can also be exported to leading third-party applications easily and quickly using the XML programming management systems such as HP OpenView, iBM Tivoli, and language. Junoscript automation tools also provide early Computer Associates Unicenter software, providing a complete, detection and automatic resolution of potential problems consolidated view of network operations. related to the operating system. QFX3000 QFabric Switch QFX3100 QFABRIC DIReCtOR QFX3500 QFABRIC NODe QFX3008 QFABRIC INteRCONNeCtQFX3000 QFabric Switch Specifications AttRIBute QFX3500 QFABRIC NODe QFX3008 QFABRIC INteRCONNeCt QFX3100 QFABRIC DIReCtOR width 17.5 in. (44.45 cm) 17.5 in. (44.45 cm) 17.5 in. (44.45 cm) Height 1.75 in. (4.45 cm), 1 U 36.75 in. (93.34 cm), 21 U 3.5 in. (8.89 cm), 2 U Depth 28 in. (71.12 cm) 32 in. (81.28 cm) 23.75 in. (60.33 cm) weight 30.75 lb (13.95 kg) 675 lb (306.17 kg) fully populated 41.2 lb (18.69 kg) Power feed (voltage) 100-240 V AC (single phase) 200-240 V AC (single phase); 100-240 V AC (single phase) 240 V (three phase) Power feed (AMP rating) 7.8 A (100-127 V) Single phase: 16 A per input 8A 3.8 A (200 -240 V) Three phase: 25.5 A per phase Power consumption (maximum) 365 watts 5,200 watts 476 watts Power consumption (nominal) 295 watts 3,000 watts 220 watts Heat dissipation (maximum) 1,250 BTU 1,775 BTU 1,624 BTU Heat dissipation (nominal) 1,010 BTU 1,024 BTU 751 BTU Air flow direction Front to back Front to back Front to back rack mount options 4-pole rack mount 4-pole rack mount 4-pole rack mount 2-pole mid-mount NEBS test/certification status 2H 2011 2H 2011 N/A Ambient temperature range 32° through 104° F 32° through 104° F 41° through 104° F (0° through 40° C) (0° through 40° C) (5° through 40° C) Operating altitude range 13,000 ft (3,962 m) 13,000 ft (3,962 m) 10,000 ft (3,048 m) 5
  6. 6. Specifications (continued)Layer 2 Features • Twelve hardware queues per port per Node (8 unicast and• VLAN—iEEE 802.1Q VLAN trunking 4 multicast)• routed VLAN interface (rVi) • Strict priority queuing (SPQ), shaped deficit weighted-round• Port-based VLAN robin (SDwrr), weighted random early detection (wrED), weighted tail drop• MAC address filtering • 802.1p remarking• Static MAC address assignment for interface • Layer 2 classification criteria—interface, MAC address,• Per VLAN MAC learning (limit) Ethertype, 802.1p, VLAN• Link aggregation and Link Aggregation Control Protocol (LACP) • Congestion avoidance capabilities—wrED (iEEE 802.3ad) • Trust iEEE 802.1p/DSCP (ingress)• iEEE 802.1AB Link Layer Discovery Protocol (LLDP) • remarking of bridged packets• Jumbo frame (9,216 bytes)Link Aggregation traffic Mirroring • Port-based• LAG load-sharing algorithm—bridged or routed (unicast or multicast) traffic • LAG port - iP—Session initiation Protocol (SiP), Dynamic internet • VLAN-based Protocol (DiP), TCP/UDP source port, TCP/UDP destination • Filter-based port • Mirror to local and remote destinations (L2 over VLAN)* - Layer 2 and non-iP—MAC source address, MAC destination Data Center Bridging (DCB) address, Ethertype, VLAN iD, source port • Priority Flow Control (PFC)—iEEE 802.1Qbb - FCoE packet—service identifier (SiD), direct inward dialing • Enhanced Transmission Selection (ETS)—iEEE 802.1Qaz (DiD), OxiD, source port • Data Center Bridging Exchange Protocol (DCBX)—part of theLayer 3 Features (IPv4) ETS standard• Static routing • Ethernet Congestion Management (QCN)—iEEE 802.1Qau*• routing policy Fibre Channel over ethernet (FCoe)• routing protocols (OSPF, BGP, iS-iS*, riP*, etc.) • FCoE transit switch (FiP snooping)• routed ports, rVis, and L3 LAG • FCoE-FC gateway• Virtual routing and forwarding (VrF): VrF-lite*, VrF-aware unicast (BGP, OSPF)* Fiber Channel Standard • Fibre Channel port speeds—2, 4, 8 GbpsMulticast Features • Fibre Channel port types—N_Port, VF_Port• iGMP snooping v1 and v2 • Fibre Channel classes of service—Class 3• iGMP snooping v3* • Fibre Channel services— N_Port iD Virtualization (NPiV)• Layer 3 multicast routing protocols* gateway• Multicast Source Directory Protocol (MSDP)* • FCoE Support—FC-BB-5 FC-BB_E including FiP Snooping• Protocol independent Multicast Version 2 (PiMv2) sparse mode (PiM-SM) and source-specific multicast (SSM)* QFabric Architecture Server Virtualization Management• Bootstrap router (BSr), auto-rP) and static rP* • Junos Space Virtual Control • iEEE 802.1Qbg*Security and Firewall Filters (ACLs)• Secure interface login and password QFabric Architecture Management and Operations• rADiUS • role-based CLi management and access• TACACS+ • CLi via console and SSH• ingress and egress firewall filters—allow and deny, port ACLs, • Show and debug commands, statistics VLAN ACLs, routed ACLs • Fabric ping and traceroute• ACL actions—logging, system logging, reject, mirror to an • Junos OS configuration rescue and rollback interface, counters, assign forwarding class, permit, drop, police, • SNMP v1/v2 and v3* mark • XML/NETCONF• SSH v1, v2 • Junoscript automation• Local proxy Address resolution Protocol (ArP) • Junos SDK• Static ArP support• Storm control, port error disable and auto-recovery• Control plane denial-of-service (DoS) protectionQuality of Service (QoS)• Layer 2 QoS—classification, rewrite, queuing• Layer 3 QoS*• rate limiting - ingress/egress policing—1 rate 2 color, 1 rate 3 color, 2 rate 3 color - Egress policer—policer mark down action - Egress shaping—per queue, per port* roadmap 6
  7. 7. Specifications (continued)QFX3000 QFabric Switch Performance and Scale Supported RFCQFX3000 QFabric Switch performance and scale can be • rFC 768 UDPcharacterized by the cumulative resources of the Nodes being • rFC 791 iP • rFC 792 iCMPused for the network design. Following are the scale numbers for • rFC 793 TCPthe QFabric architecture: • rFC 826 ArP• Platform: • rFC 894 iP over Ethernet - QFabric Directors: 2 • rFC 903 rArP - QFabric interconnects: 4 • rFC 1027 Proxy ArP - QFabric Nodes: 128 • rFC 1058 riP v1• Layer 2 • rFC 1112 iGMP v1 - MAC addresses: 128,000 through 1,536,000 • rFC 1122 Host requirements - VLANs: 4,096 • rFC 1195 Use of Open Systems interconnection (OSi) iS-iS for - Number of LAGs: 48 per QFNode routing in TCP/iP and Dual Environments (TCP/iP transport - Number of ports per LAG: 8 only) - LLDP peers: 348 • rFC 1256 iPv4 iCMP router Discovery (irDP) - Jumbo frame : 9,216 bytes • rFC 1492 TACACS+• Layer 3 • rFC 1519 Classless interdomain routing (CiDr) - rVis: 2,000 • rFC 1587 OSPF NSSA Option - iPv4 unicast routes: 16,000 for QFabric architecture • rFC 1591 Domain Name System (DNS) - OSPF neighbors: 256 • rFC 1745 BGP4/iDrP for iP-OSPF interaction - BGP peers: 256 • rFC 1765 OSPF Database Overflow• Multicast • rFC 1771 Border Gateway Protocol 4 - Multicast groups: 3,900 for QFabric architecture • rFC 1812 requirements for iP Version 4 routers• QoS • rFC 1965 Autonomous System Confederations for BGP - Policers (ingress and egress): 1.500 per QFabric Node • rFC 1997 BGP Communities Attribute - Queues: 12 per QFabric Node, 8 unicast and 4 multicast • rFC 2030 SNTP, Simple Network Time Protocol• Security • rFC 2138 rADiUS Authentication - Firewall filters (ACLs): 1,500 per QFabric Node • rFC 2139 rADiUS Accounting - Traffic mirroring • rFC 2154 OSPF with Digital Signatures (Password, MD-5) • Mirroring destination ports per QFabric Node: 4 • rFC 2236 iGMP v2 • Mirroring destination VLANs per QFabric Node: 256 • rFC 2267 Network ingress Filtering• FCoE/FC • rFC 2328 OSPF v2 (Edge-mode) - FCoE interfaces: 6,144 maximum • rFC 2362 PiM-SM (Edge-mode) - FC interfaces: 1,536 maximum • rFC 2370 OSPF Opaque LSA Option - FC sessions: 3,000 per QFabric Node • rFC 2385 TCP MD5 Authentication for BGPv4 - VF_Ports: 6,144 maximum • rFC 2439 BGP route Flap Damping • rFC 2453 riP v2QFabric Architecture Standards Compliance • rFC 2474 DiffServ Precedence, including 8 queues/portIeee Standard • rFC 2475 DiffServ Core and Edge router Functions• iEEE 802.1AB: Link Layer Discovery Protocol (LLDP) • rFC 2597 DiffServ Assured Forwarding (AF)• iEEE 802.1p: Class-of-service (CoS) prioritization • rFC 2598 DiffServ Expedited Forwarding (EF)• iEEE 802.1Q-2006: VLAN tagging • rFC 2796 BGP route reflection (supersedes rFC 1966)• iEEE 802.3ab: 1000BASE-T • rFC 2918 route refresh Capability for BGP-4• iEEE 802.3z: 1000BASE-X • rFC 3376 iGMP v3• iEEE 802.3ae: 10-Gigabit Ethernet • rFC 3392 Capabilities Advertisement with BGP-4• iEEE 802.3ad: Link Aggregation Control Protocol (LACP) • rFC 3569 Draft-ietf-ssm-arch-06.txt PiM-SSM PiM Source• iEEE 802.1Qbb Specific Multicast• iEEE 802.1Qaz • rFC 3623 OSPF Graceful restart• iEEE 802.1Qau* • •rFC 4360 BGP Extended Communities Attribute• iEEE 802.1Qbg* • rFC 4486 Subcodes for BGP Cease Notification message • Draft-ietf-idr-restart-10.txt: Graceful restart Mechanism fort11 Standards: BGP• iNCiTS T11 FC-BB-5 • Draft-ietf-isis-restart-02: restart Signaling for iS-iS • PiM-DM Draft iETF PiM: Dense Mode draft-ietf-idmr-pimdm-05. txt, draft-ietf-pim-dm-new-v2-04.txt* roadmap 7
  8. 8. Specifications (continued) Ordering InformationSupported MIBs MODeL NuMBeR DeSCRIPtION• rFC 1155 Structure and identification of Management QFX3500 Node Base Hardware information for TCP/iP-based internets• rFC 1157 A Simple Network Management Protocol (SNMP) QFX3500-48S4Q-ACr QFX3500, 48 SFP+/SFP and 4 QSFP ports, redundant dual AC power supply, front-to-back air• rFC1212 Concise MiB definitions flow• rFC 1213 Management information Base for Network Management of TCP/iP-Based internets: MiB-ii (Partial) QFX3500 Node Hardware Spare• rFC 1215 A Convention for Defining Traps for use with the SNMP QFX3500-48S4Q QFX3500, 48 SFP+/SFP and 4 QSFP ports (spare)• rFC 1901 introduction to Community-based SNMPv2 QFX3500-MB Management board for QFX3500-48S4Q (spare)• rFC 1905 Protocol Operations for Version 2 of the Simple QFX3500-FANAi Fan module (intake) for QFX3500-48S4Q (spare) Network Management Protocol (SNMPv2) QFXC01-PwrACi-650A 650 w power supply (with intake air flow) (spare)• rFC 1907 Management information Base for Version 2 of the Simple Network Management Protocol (SNMPv2) QFX3500-rMAr rear rack mount bracket A (spare)• rFC 2011 SNMPv2 for internet protocol using SMiv2 QFX3008 Interconnect Base Hardware• rFC 2012 SNMPv2 for transmission control protocol using SMiv2 QFX3008-CHASA-BASE QFX3008, 8 slots A/C base system with redundant• rFC 2013 SNMPv2 for user datagram protocol using SMiv2 dual control card, six redundant power supplies• rFC 2233 The interfaces Group MiB Using SMiv2 QFX3008-SF16Q Front card with 16-port QSFP• rFC 2571 An Architecture for describing SNMP Management Frameworks (read-only access) (SNMPv3*) QFX3008 Interconnect Hardware Spares• rFC 2572 Message Processing and Dispatching for the SNMP QFXC08-ACTrAY-C19 A/C power wiring tray (single phase) (read-only access) (SNMPv3*) QFXC08-ACTrAY-D A/C power wiring tray (three phase delta)• rFC 2576 Coexistence between SNMP Version 1, Version 2, and Version 3 (SNMPv3*) QFXC08-ACTrAY-w A/C power wiring tray (three phase wYE)• rFC 2578 SNMP Structure of Management information MiB QFXC08-CABMAN Cable management module• rFC 2579 SNMP Textual Conventions for SMiv2 QFXC08-DOOr Front door• rFC 2580 Conformance Statements for SMiv2 QFXC08-CHASA QFX3008 chassis, 8 slots A/C• rFC 2863 interface Group MiB• rFC 3410 introduction and Applicability Statements for internet QFXC08-CB4S Control card with 4x1/10GbE SFP+ ports Standard Management Framework (SNMPv3*) QFXC08-PwrAC-4000 4,000 w A/C PSU (spare)• rFC 3411 An architecture for describing SNMP Management QFXC08-FANT Top fan tray (spare) Frameworks (SNMPv3*) QFXC08-FANB Bottom fan tray (spare)• rFC 3412 Message Processing and Dispatching for the SNMP (SNMPv3*) QFXC08-FANS Side fan tray (spare)• rFC 3413 Simple Network Management Protocol (SNMP) (all QFXC08-FBLNK Front slot blank cover MiBs are supported except the Proxy MiB) (SNMPv3*) QFXC08-FFLTr Front air filter (spare)• rFC 3415 View-based Access Control Model (VACM) for the SNMP QFXC08-SFLTr Side air filter (spare)• rFC 3416 Version 2 of the Protocol Operations for the SNMP QFX3008-Sr1 rear fabric card (spare)• rFC 3417 Transport Mappings for the SNMP QFX3100 Director Base Hardware• rFC 3418 Management information Base (MiB) for the SNMP QFX3100-GBE-ACr QFX3100 base system with redundant AC power (SNMPv3*) supply, dual disk and network interface cards• rFC 3584 Coexistence between Version 1, Version 2, and Version 3* of the internet-standard Network Management Framework QFX3100 Director Hardware Spares• Support for the following Juniper Networks enterprise-specific QFX3100-NM-4GE Network interface card with 4x10/100/1000BASE- MiBs: TX (spare) - Chassis MiB (jnx-chassis.mib) QFX3100-PwrAC- 560 w power supply A (spare) - interface MiB (jnx-if-extensions.mib) 560A - Power Supply Unit MiB (jnx-power-supply-unit.mib) QFX3100-HDD-2TB 2 TB hard drive (spare) QFX3100-FANA Fan module (spare)Juniper Networks Services and SupportJuniper Networks is the leader in performance-enabling servicesthat are designed to accelerate, extend, and optimize yourhigh-performance network. Our services allow you to maximizeoperational efficiency while reducing costs and minimizingrisk, achieving a faster time to value for your network. JuniperNetworks ensures operational excellence by optimizing thenetwork to maintain required levels of performance, reliability, andavailability. For more details, please visit www.juniper.net/us/en/products-services.* roadmap 8
  9. 9. Ordering Information (continued) Services and Support MODeL NuMBeR DeSCRIPtION Juniper Services for the QFabric Architecture Juniper Networks QFabric architecture comes with the Optics and transceivers essential services needed to assist customers across the entire QFX-QSFP-40G-Sr4 QSFP+ 40GBASE-Sr4 40-Gigabit Optics, 850 nm for up to 150 m transmission on multimode fiber- QFabric family of products lifecycle, from planning, design and optic (MMF) implementation, through operation and optimization phases. QFX-SFP-10GE-Sr SFP+ 10GBASE-Sr 10-Gigabit Ethernet Optics, 850 These services can quickly “operationalize” deployments, while nm for up to 300 m transmission on MMF minimizing business disruptions and optimizing the QFabric QFX-SFP-10GE-USr SFP+ 10-Gigabit Ethernet Ultra Short reach Optics, archtecture’s performance. They are designed to mitigate the risk 850 nm for 10 m on OM1, 20 m on OM2, 100 m on OM3 MMF of migrating from legacy technologies, while accelerating revenue QFX-SFP-10GE-Lr SFP+ 10GBASE-Lr 10-Gigabit Ethernet Optics, generating activities and delivering value from the beginning. 1,310 nm for 10 km transmission on single mode These required services include: fiber-optic (SMF) • QFabric Architecture Start-Up Service: Provides a choice of QFX-SFP-DAC-1M SFP+ 10-Gigabit Ethernet Direct Attach Copper (twinax copper cable) 1 m design review levels to ensure that deployments leverage the QFX-SFP-DAC-3M SFP+ 10-Gigabit Ethernet Direct Attach Copper advantages of QFabric architecture and are ready for a smooth (twinax copper cable) 3 m system bring up, which is included in the service. QFX-SFP-DAC-5M SFP+ 10-Gigabit Ethernet Direct Attach Copper • QFabric Architecture resident Engineer: Juniper’s deep (twinax copper cable) 5 m technical network design and operational experience can be QFX-SFP-DAC-1MA SFP+ 10-Gigabit Ethernet Direct Attach Copper quickly leveraged by customers, working in daily proximity with a (active twinax copper cable) 1 m resident Engineer. QFX-SFP-DAC-3MA SFP+ 10-Gigabit Ethernet Direct Attach Copper (active twinax copper cable) 3 m • Juniper Care and Juniper Care Plus: QFX-SFP-DAC-5MA SFP+ 10-Gigabit Ethernet Direct Attach Copper - Juniper Care offers best-in-class 24x7 support, rated #1 for (active twinax copper cable) 5 m Mission Critical Hardware Support in consecutive years1. QFX-SFP-DAC-7MA SFP+ 10-Gigabit Ethernet Direct Attach Copper - Juniper Care also delivers award-winning online support, (active twinax copper cable) 7 m rated among the 10 best for five consecutive years 2—a record QFX-SFP-DAC-10MA SFP+ 10-Gigabit Ethernet Direct Attach Copper unequaled in the industry. (active twinax copper cable) 10 m - Junos Space Service Now, included with all Juniper Care QFX-SFP-8GFC-Sw SFP 2/4/8-Gbps Fibre Channel Sw Optics contracts, helps streamline operations and reduce mean QFX-SFP-1GE-T SFP 1000BASE-T 10/100/1000 Copper Transceiver Module for up to 100 m transmission on Cat5 time to recovery (MTTr) through its automated monitoring and diagnostic capabilities. Asset management capabilities QFX-SFP-1GE-SX SFP 1000BASE-SX Gigabit Ethernet Optics, 850 nm for up to 550 m transmission on MMF help increase productivity by automating time-consuming QFX-SFP-1GE-LX SFP 1000BASE-LX Gigabit Ethernet Optics, 1,310 manual tasks. nm for 10 km transmission on SMF - Juniper Care Plus for QFabric architecture keeps the new Software Licenses network at optimum readiness by leveraging Juniper’s service QFX3000-JSL-EDGE- QFX3000 QFabric Switch QFabric Node feature automation platform—Junos Space Service insight—which FAB license delivers targeted actionable information on a proactive basis. QFX-JSL-DrCTr-ADV1 QFabric architecture advanced feature license for Accelerated Expert To Expert Access and an account assigned iS-iS, BGP, and iPv6 routing Service Manager are also included. Consulting credits which QFX3008-JSL-DrCTr- QFX3008 base QFabric software are easily leveraged for proactive services tuned to your FAB QFabric technology environment, and Juniper training credits QFX-JSL-DrCTr-FC QFabric architecture feature license for FCoE-to-FC gateway to accelerate knowledge transfer to your own organization are QFX-JSL-DrCTr- QFabric architecture feature license for FCoE-to-FC also included. FC-C16 gateway (capacity 16) About Juniper Networks Juniper Networks is in the business of network innovation. From devices to data centers, from consumers to cloud providers, Juniper Networks delivers the software, silicon and systems that transform the experience and economics of networking. The company serves customers and partners worldwide. Additional information can be found at www.juniper.net.Technology Services industry Association (TSiA)1Association of Support Professionals (ASP). No other networking company has won more than two consecutive ASP awards.2 9
  10. 10. 10
  11. 11. 11
  12. 12. Corporate and Sales Headquarters APAC Headquarters eMeA Headquarters To purchase Juniper Networks solutions,Juniper Networks, inc. Juniper Networks (Hong Kong) Juniper Networks ireland please contact your Juniper Networks1194 North Mathilda Avenue 26/F, Cityplaza One Airside Business Park representative at 1-866-298-6428 orSunnyvale, CA 94089 USA 1111 King’s road Swords, County Dublin, ireland authorized reseller.Phone: 888.JUNiPEr (888.586.4737) Taikoo Shing, Hong Kong Phone: 35.31.8903.600or 408.745.2000 Phone: 852.2332.3636 EMEA Sales: 00800.4586.4737Fax: 408.745.2100 Fax: 852.2574.7803 Fax: 35.31.8903.601www.juniper.netCopyright 2011 Juniper Networks, inc. All rights reserved. Juniper Networks, the Juniper Networks logo, Junos,NetScreen, and ScreenOS are registered trademarks of Juniper Networks, inc. in the United States and othercountries. All other trademarks, service marks, registered marks, or registered service marks are the property oftheir respective owners. Juniper Networks assumes no responsibility for any inaccuracies in this document. JuniperNetworks reserves the right to change, modify, transfer, or otherwise revise this publication without notice.1000393-002-EN Sept 2011 Printed on recycled paper 12