Stackwise/Stackwise Plus | Device Availability Technologies

Cisco StackWise and StackWise Plus technologies create a unified, logical switching architecture through the linkage of multiple, fixed configuration 3750G and 3750E switches.

Note

Cisco 3750G switches use StackWise technology, and Cisco 3750E switches can use either StackWise or StackWise Plus. (StackWise Plus is used only if all switches within the group are 3750E switches; whereas, if some switches are 3750E and others are 3750G, StackWise technology will be used.) Also to prevent excessive wordiness, “StackWise” is used in this section to refer to both StackWise and StackWise Plus technologies, with the exception of explicitly pointing out the differences between the two at the end of this section.

Cisco StackWise technology intelligently joins individual switches to create a single switching unit with a 32-Gbps switching stack interconnect. Configuration and routing information is shared by every switch in the stack, creating a single switching unit. Switches can be added to and deleted from a working stack without affecting availability.

The switches unite into a single logical unit using special stack interconnect cables that create a bidirectional closed-loop path. This bidirectional path acts as a switch fabric for all the connected switches. Network topology and routing information is updated continuously through the stack interconnect. All stack members have full access to the stack interconnect bandwidth. The stack is managed as a single unit by a master switch, which is elected from one of the stack member switches.

Each switch in the stack has the capability to behave as a master in the hierarchy. The master switch is elected and serves as the control center for the stack. Each switch is assigned a number. Up to nine separate switches can be joined together.

Each stack of Cisco Catalyst 3750 Series switches has a single IP address and is managed as a single object. This single IP management applies to activities such as fault detection, VLAN creation and modification, security, and quality of service (QoS) controls. Each stack has only one configuration file, which is distributed to each member in the stack. This allows each switch in the stack to share the same network topology, MAC address, and routing information. In addition, it allows for any member to immediately take over as the master, if there is a master failure.

To efficiently load balance the traffic, packets are allocated between two logical counter-rotating paths. Each counter-rotating path supports 16 Gbps in both directions, yielding a traffic total of 32 Gbps bidirectionally. When a break is detected in a cable, the traffic is immediately wrapped back across the single remaining 16-Gbps path (within microseconds) to continue forwarding.

Switches can be added and deleted to a working stack without affecting stack availability. (However, adding additional switches to a stack might have QoS performance implications) Similarly, switches can be removed from a working stack without any operational effect on the remaining switches.

Stacks require no explicit configuration but are automatically created by StackWise when individual switches are joined together with stacking cables, as shown in Figure 1. When the stack ports detect electromechanical activity, each port starts to transmit information about its switch. When the complete set of switches is known, the stack elects one of the members to be the master switch, which will be responsible for maintaining and updating configuration files, routing information, and other stack information. This process is referred to as hot stacking.

 
Figure 1: Catalyst 3750G StackWise cabling

Courtesy of Cisco Systems, Inc. Unauthorized use not permitted.

Note

Master switch election occurs only on stack initialization or if there is a master switch failure. If a new, more favorable switch is added to a stack, this will not trigger a master switch election, nor will any sort of preemption occur.

Each switch in the stack can serve as a master, creating a 1:N availability scheme for network control. In the unlikely event of a single unit failure, all other units continue to forward traffic and maintain operation. Furthermore, each switch is initialized for routing capability and is ready to be elected as master if the current master fails. Subordinate switches are not reset so that Layer 2 forwarding can continue uninterrupted.

The three main differences between StackWise and StackWise Plus are as follows:

• StackWise uses source stripping, and StackWise Plus uses destination stripping (for unicast packets). Source stripping means that when a packet is sent on the ring, it is passed to the destination, which copies the packet and then lets it pass all the way around the ring. After the packet has traveled all the way around the ring and returns to the source, it is stripped off of the ring. This means bandwidth is used up all the way around the ring, even if the packet is destined for a directly attached neighbor. Destination stripping means that when the packet reaches its destination, it is removed from the ring and continues no further. This leaves the rest of the ring bandwidth free to be used. Thus, the throughput performance of the stack is multiplied to a minimum value of 64 Gbps bidirectionally. This capability to free up bandwidth is sometimes referred to as spatial reuse.

  Note

  Even in StackWise Plus, broadcast and multicast packets must use source stripping because the packet might have multiple targets on the stack.

• StackWise Plus can locally switch; StackWise cannot. In StackWise Plus, packets originating and destined to ports on the same local switch will not have to traverse the Stack ring, which results in more efficient switching. In contrast, in StackWise, because there is no local switching and because there is source stripping, even locally destined packets must traverse the entire stack ring.
• StackWise Plus can support up to two Ten Gigabit Ethernet ports per Cisco Catalyst 3750-E.

Finally, both StackWise and StackWise Plus can support Layer 3 NSF when two or more nodes are present in a stack. NSF is discussed in the following section, along with SSO.

Device Availability Technologies

Most network designs have single points of failure, and the overall availability of the network might be dependent on the availability of a single device. A prime example of this is the access layer of a campus network. Most endpoint devices connect to the access switch through a single network interface card (NIC); this is referred to as being single-homed; therefore, access switches represent a single point of failure for all attached single-homed devices, including CTS codecs.

Note

Beginning with CTS 1.5 software, Cisco TelePresence Multipoint Switches can utilize a NIC teaming feature that can enable these to be multihomed devices, that is, devices that connect to multiple access switches. Multihoming eliminates the access switch from being a single-point of failure and thus improves overall availability.

Ensuring the availability of the network services is often dependent on the resiliency of the individual devices. Device resiliency, as with network resiliency, is achieved through a combination of the appropriate level of physical redundancy, device hardening, and supporting software features. Studies indicate that most common failures in campus networks are associated with Layer 1 failures, from components such as power supplies, fans, and fiber links. The use of diverse fiber paths with redundant links and linecards, combined with fully redundant power supplies and power circuits, are the most critical aspects of device resiliency. The use of redundant power supplies becomes even more critical in access switches with the introduction of Power over Ethernet (PoE) devices such as IP phones. Multiple devices are now dependent on the availability of the access switch and its capability to maintain the necessary level of power for all the attached end devices. After physical failures, the most common cause of device outage is often related to the failure of supervisor hardware or software. The network outages due to the loss or reset of a device due to supervisor failure can be addressed through the use of supervisor redundancy. Cisco Catalyst switches provides two mechanisms to achieve this additional level of redundancy:

• Cisco StackWise/StackWise-Plus
• Cisco Nonstop Forwarding (NSF) with Stateful Switchover (SSO)

Both of these mechanisms, discussed in the following sections, provide for a hot active backup for the switching fabric and control plane, thus ensuring that data forwarding and the network control plane seamlessly recover (with subsecond traffic loss, if any) during any form of software or supervisor hardware crash.

TelePresence Phases of Deployment

As TelePresence technologies evolve, so too will the complexity of deployment solutions. Therefore, customers will likely approach their TelePresence deployments in phases, with the main phases of deployment as follows:

• Phase 1. Intracampus/Intra-enterprise deployments: Most enterprise customers will likely begin their TelePresence rollouts by provisioning (point-to-point) intra-enterprise TelePresence deployments. View this model as the basic TelePresence building block on which you can add more complex models.
• Phase 2. Intra-enterprise multipoint deployments: Because collaboration requirements might not always be facilitated with point-to-point models, the next logical phase of TelePresence deployment is to introduce multipoint resources to the intra-enterprise deployment model. Phases 1 and 2 might at times be undertaken simultaneously.
• Phase 3. Intercompany deployments: To expand the application and business benefits of TelePresence meetings to include external (customer- or partner-facing) meetings, an intercompany deployment model can be subsequently overlaid over either point-to-point or multipoint intra-enterprise deployments.
• Phase 4. TelePresence to the executive home: Because of the high executive-perk appeal of TelePresence and the availability of high-speed residential bandwidth options (such as fiber to the home), some executives might benefit greatly from deploying TelePresence units to their residences. Technically, this is simply an extension of the intra-enterprise model but might also be viewed as a separate phase because of the unique provisioning and security requirements posed by such residential TelePresence deployments, as illustrated in Figure 1.
  
  
  
   
  Figure 1: TelePresence to the executive home (an extension of the intra-enterprise deployment model)