Summary of the content on the page No. 1
Integrating the Cisco Catalyst Blade
Switch 3020 for the HP c-Class
BladeSystem into the Cisco Data
Center Network Architecture
Design Guide
© 2008 Cisco Systems, Inc. All rights reserved. This document is Cisco Public Information.
Summary of the content on the page No. 2
Design Guide Contents Introduction ..................................................................................................................................... 3 HP c-Class BladeSystem Enclosure Overview............................................................................. 3 Cisco Catalyst Blade Switch 3020 for HP...................................................................................... 5 Cisco Catalyst Blade Switch 3020 Features ...............................
Summary of the content on the page No. 3
Design Guide Introduction ® ® This guide provides best design practices for deploying the Cisco Catalyst Blade Switch 3020 for the HP c-Class BladeSystem enclosure within the Cisco Data Center Networking Architecture. It describes the internal components of the blade-server enclosure and Cisco Catalyst Blade Switch 3020 and explores different methods of deployment. HP c-Class BladeSystem Enclosure Overview The HP c-Class BladeSystem enclosure represents the next generation of blade-se
Summary of the content on the page No. 4
Design Guide Figure 2. Enclosure Interconnections Using Full-Height Servers Figure 3. Enclosure Interconnections Using Half-Height Servers © 2008 Cisco Systems, Inc. All rights reserved. This document is Cisco Public Information. Page 4 of 28
Summary of the content on the page No. 5
Design Guide Cisco Catalyst Blade Switch 3020 for HP This section briefly describes the Cisco Catalyst Blade Switch 3020 for HP and explains how the blade servers within the HP c-Class BladeSystem are physically connected to the switching modules. The Cisco Catalyst Blade Switch 3020 provides enhanced Layer 2 services (known as Layer 2+ or Intelligent Ethernet switching) to the HP c-Class BladeSystem. The Cisco Catalyst Blade Switch 3020 enhances basic Layer 2 switching by including Cis
Summary of the content on the page No. 6
Design Guide Cisco Catalyst Blade Switch 3020 Features This section highlights information about the protocols and features provided by the Cisco Catalyst Blade Switch 3020 that help integrate the HP c-Class BladeSystem enclosure into the Cisco Data Center Network Architecture. Spanning Tree The Cisco Catalyst Blade Switch 3020 supports different versions of the Spanning Tree Protocol and associated features, including the following: ● Rapid Spanning Tree Protocol (RSTP), based on IEE
Summary of the content on the page No. 7
Design Guide A scenario wherein BPDUs are lost may be caused by unidirectional links, which can cause Layer 2 loops. To prevent this problem, use Loop Guard and UDLD. Loop Guard prevents a port from forwarding as a result of missed BPDUs, which might cause a Layer 2 loop that could bring down the network. UDLD allows devices to monitor the physical configuration of fiberoptic or copper Ethernet cables and detect when a unidirectional link exists. When a unidirectional link is detected,
Summary of the content on the page No. 8
Design Guide Note: The IEEE 802.1w protocol is enabled by default when running spanning tree in RPVST+ or MST mode on the Cisco Catalyst Blade Switch 3020. The Cisco Catalyst Blade Switch 3020 enables PVST+ for VLAN 1 by default. The Spanning Tree Protocol uses the path cost value to determine the shortest distance to the root bridge. The port path cost value represents the media speed of the link and is configurable on a per-interface basis, including Cisco EtherChannel interfaces. To
Summary of the content on the page No. 9
Design Guide Figure 5. RSPAN Example Link Aggregation Protocols Cisco Fast EtherChannel interfaces and Gigabit EtherChannel interfaces are logically bundled, and they provide link redundancy and scalable bandwidth between network devices. PAgP and LACP help automatically create these channels by exchanging packets between Ethernet interfaces and negotiating a logical connection. PAgP is a Cisco proprietary protocol that can be run only on Cisco switches or on switches manufactured b
Summary of the content on the page No. 10
Design Guide Figure 6. Alternative Network Configuration Data Center Network Architecture The architecture of the data center infrastructure must address the requirements necessary to create a highly available, scalable, and secure network. This section describes the basic architecture necessary to meet these goals. It is a synopsis of the Cisco Data Center Network Architecture; for details about this architecture, visit: http://www.cisco.com/en/US/solutions/ns340/ns517/ns224/ns304/
Summary of the content on the page No. 11
Design Guide Figure 7. Data Center Front-End Network Aggregation Layer The aggregation layer is a point of convergence for network traffic that provides connectivity between server farms and the rest of the enterprise. The aggregation layer supports Layer 2 and Layer 3 functions and presents an ideal location for deploying centralized application, security, and management services. Shared across the access layer server farms, these data center services provide an efficient, scalable
Summary of the content on the page No. 12
Design Guide Layer 2 adjacency in the server farm allows for the deployment of servers or clusters that require the exchange of information done at Layer 2 only. It also readily supports access to network services in the aggregation layer such as load balancers and firewalls, enabling an efficient use of shared, centralized network services by the server farms. In contrast, if services are deployed at each access switch, the benefit of those services is limited to the servers directly att
Summary of the content on the page No. 13
Design Guide link redundancy combined with a deterministic topology design to achieve application-availability requirements. Servers are typically configured with multiple NICs and dual homed to the access layer switches to provide backup connectivity to the business application. High availability is an important design consideration in the data center. The Cisco Catalyst Blade Switch 3020 has numerous features and characteristics that contribute to a reliable, highly available network.
Summary of the content on the page No. 14
Design Guide ● Network Fault Tolerance (NFT) ● Transmit Load Balancing (TLB) ● Switch Assisted Load Balancing (server load balancing [SLB]) NFT teaming creates a virtual interface by grouping the blade-server network adapters into a team. One adapter is the primary active interface and all other adapters are in a standby state. The virtual adapter uses a single MAC address and a single Layer 3 address. NFT provides adapter fault tolerance by monitoring the state of each team member’s
Summary of the content on the page No. 15
Design Guide Slot Count The data center infrastructure must be flexible enough to allow growth in both server capacity and service performance. Connecting a blade system directly into the aggregation layer places more significance on the number of slots available to accommodate blade-system uplinks and integrated services. Traditionally, the access layer provides the port density necessary to allow the physical growth of server farms. Modular access layer switches offer connectivity to
Summary of the content on the page No. 16
Design Guide The data center core is a mechanism to replicate and horizontally scale the data center environment. In the recommended design the aggregation and access layers are regarded as a module that can be duplicated to extend the enterprise. Each data center module provides its own network services locally in the aggregation switches. This approach allows the network administrator to determine the limits of each data center module and replicate as necessary. Figure 9 depicts the d
Summary of the content on the page No. 17
Design Guide The Cisco Catalyst Blade Switch 3020 contains an additional Fast Ethernet port, which connects to the HP c-Class BladeSystem Onboard Administrator, providing OOB management using the insight manager interface. The user may also use this path to access the CLI functions of the switch, transfer SNMP information, and upload software images and configuration files. This path is independent of the switch fabric. This Fast Ethernet port defaults to a Dynamic Host Configuration Pro
Summary of the content on the page No. 18
Design Guide Management Options The Cisco Catalyst Blade Switch 3020 switch is manageable with the following methods: ● HTTP-based device-manager GUI ● SNMP-based management applications ● Cisco IOS Software CLI The embedded device manager on the Cisco Catalyst Blade Switch 3020 provides a GUI to configure and monitor the switch through a Web browser. This scenario requires using either in- band or out-of-band management and enabling the HTTP or HTTPS server on the switch. The HTTP s
Summary of the content on the page No. 19
Design Guide Network Topologies Using the Cisco Catalyst Blade Switch 3020 This section discusses the following physical topologies: ● Recommended topology: Classic V-shaped topology with Spanning Tree Protocol ● Alternative topology: Square topology with Spanning Tree Protocol These network designs emphasize high availability in the data center by eliminating any single point of failure and by providing deterministic traffic patterns and predictable behavior during times of network c
Summary of the content on the page No. 20
Design Guide ● Primary root switch failure and recovery ● Secondary root switch failure and recovery These tests revealed the intricacies of fast convergence in the data center and the necessity for a holistic approach to high availability. Test cases that did not involve the failure of the active HSRP aggregation switch resulted in an average failover time of about 1 second. Failing the active HSRP device requires convergence at Layer 3 and resulted in a recovery time that reflected the
