Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

Navigation
Guide That Contains This Content
[+] Expand All
[-] Collapse All

    Planning the SSR Cluster Topology

    The topology of all SSR clusters begins with the four C Series Controllers required to implement the SSR Starter Kit which licenses you for two C Series Controllers configured as client nodes, each hosting a management server and SSR client components; and two C Series Controllers, each hosting a data node. Figure 1 shows the SSR Starter Kit basic configuration.

    Figure 1: Basic SSR Starter Kit Cluster

    Basic SSR Starter Kit Cluster

    Small-Scale Deployments

    For small-scale deployments, the minimum SSR configuration requires three C Series Controllers set up in the two-shared-data-node geometry. In this configuration, there are two data-client-nodes and two management servers. One management server must be running on the third C Series Controller and the other one can reside on any data-client-node. You also need to run redundant SAEs, SICs, and other SRC components such as NIC to maintain redundancy. This solution requires you to configure the data node memory size because memory is shared among all components.

    Identifying the C Series Controllers in the SSR Cluster

    Each node in the cluster is identified by its IP address. The type of node can be either a data node, a client node, or a data client node. Data-client nodes can be used only in two-shared-data-node geometry, or for demo purposes. Up to four data nodes and twenty-four client nodes can be configured. The number of data nodes configured must match the template selected with the geometry option (two-data-node, four-data-node, or two-shared-data-node).

    • Data node—C Series Controller running an instance of the data node component.
    • Client node—C Series Controller running an instance of a Juniper Networks component that is a client of the SSR (for example, SAE or SIC). In addition, each client node hosts an instance of the database front-end component (SQL node component). Up to three client nodes may optionally be configured to run an instance of the management server component.
    • Data-client-node—C Series Controller hosting a data node component, a client node component, and a management server. Data-client-nodes are used only for small-scale deployments using the two-shared-data-node cluster geometry, or for demonstration purposes when the cluster geometry is set to all-in-one.

    Note: All data nodes must have equal processor power, memory space, and available bandwidth because they are tightly coupled and share data. If the overall throughput of the data nodes varies from node to node, performance degrades. Therefore, all data nodes must be of the same C Series Controller model—for example, all C4000 models.

    An example of node ID assignments is shown in Table 1. In this example, the cluster consists of two client nodes, each hosting an SSR client component, and two data nodes.

    Note: Because Ethernet bonding is recommended, only one IP address is used for each C Series Controller.

    Table 1: Example Allocation of Node IDs

    IP Address and Type

    Bonded IP Address

    Component

    Node ID Assigned to Component

    192.168.0.18 (Data node)

    192.168.0.18

    data component

    18

    192.168.0.19 (Data node)

    192.168.0.19

    data component

    19

    192.168.0.1 (Client node)

    192.168.0.1

    Management server component

    Database front-end component (SQL component)

    Client component

    1

    6

    41

    192.168.0.2 (Client node)

    192.168.0.2

    Management server component

    Database front-end component (SQL component)

    Client component

    2

    7

    42

    Published: 2014-12-10