CA Performance

Welcome to the CA Performance Management Version 3. 2 video Configure Monitoring. Considering the diversity of modern networks, you cannot adopt a one-size-fits-all strategy for pulling devices and interfaces. Some devices and interfaces need to be pulled at different rates, w You may also want to monitor different aspects of a device, such as hard drives and environmental components. A core feature of CA Performance Management is the ability to effectively monitor at very large scale. The product arc CA Performance Management is designed with the features and flexibility required to handle these diverse monitoring requirements. Monitoring configuration involves four different item types, metric families, vendor certifications, collections, and monitoring profiles. The four different item types work together to specify CA Performance Management Monitoring behavior. T 2. First, let's take a look at metric families. Metric families allow you to group the metrics you want to pull from your monitor environment into logical, convenient sets. Metric families are diverse, supplying metric coverage for wide-ranging categories. They are vendor agnostic, allowing the user to administer monitoring in portions and not metric by metric. CA Performance Management s In most cases, the out-of-the-box metric family support is sufficient for comprehensively monitoring an environment. Here, we're viewing the metrics of the interface metric family. Out-of-the-box metric families can be monitored, or custom metric families can be created to cover new technologies or component types. Monitoring profiles, which is covered later Next, let's take a look at vendor certifications. Metric families require vendor certifications, w Vendor certifications specify the vendor-specific manner in w Vendor certifications define data acquisition and metric calculations for proprietary vendor MIB objects and MIB objects from standard RFCs. Not every metric in a metric family has an evaluation expression and a vendor certification. Support for metrics often vary widely from vendor to vendor or from vendors to RFC specification. Vendor certifications allow you to encapsulate vendor or device-specific calculations on a per metric family basis. C performance management s The out-of-the-box vendor certification functionality enables monitoring a wide range of common devices that exist in the customer's environment. Out-of-the-box vendor certifications can be modified or custom ones can be created in cases where the out-of-the-box coverage is insufficient. There may be times you want to change the vendor certification priority order for a metric family. Changing the priority triggers an automatic metric family update. If a new Changing priority is often done to ensure a new vendor-specific certification containing greater metric coverage takes precedence over a certification, often an RFC standard, w Now let's review collections. To efficiently pull network devices, the device must be organized into special groups called collections. A collection is a group of devices to w A set of metric families is associated with each device in a collection. These associations occur when a set of monitoring parameters, w Each set of monitoring parameters is called a monitoring profile, w Devices can belong to multiple collections. Collection members Groups used for report organization are different from collections. The data aggregator comes with a set of out-of-the-box collections. The collections are read-only and are automatically updated as devices are added, deleted, or updated. The out-of-the-box DA collections are used in the s It is good practice to create custom collections when extending the s Custom collections are created in CA performance center and then synchronized onto the data aggregator. Group members For example, when a router is added to a collection, its interfaces are added as well when t Group members Rule-based members Dynamic members For example, here we're creating a dynamic group with rules to include servers in Boston. Any new Boston servers will be added to the collection automatically, and any servers moved from the Boston location will be automatically deleted from the collection. The rules defining custom group members These supply to user with the ability to define the group members Finally, let's take a look at monitoring profiles. A monitoring profile is a set of parameters w Monitoring profiles consist of a list of metric families, a metric poll rate, change detection configuration, and event rules. Collections are assigned to monitoring profiles. Monitoring profiles enable metric polling using specified monitoring characteristics. The association of a collection with a monitoring profile is the key action w The profile settings are applied to each device in an assigned collection. A collection can be assigned to multiple monitoring profiles. Each metric family in a profile's list is associated with each device in an assigned collection. T Honoring profiles determine the content and frequency of poll data, event behavior, and device component update behavior. In addition to a collection being assigned to multiple monitoring profiles, a monitoring profile can have multiple collections associated with it. CA Performance Management s In most cases, where out-of-the-box coverage is insufficient, CA Performance Management allows for the creation of custom monitoring profiles. Custom monitoring profiles can be created from scratch or from copying and modifying an existing profile. You may need a custom monitoring profile to change the polling interval of an out-of-the-box monitoring profile. Add event support, such as legacy event rules contained wit Add or change filtering on a metric family or incorporate a newly created custom metric family. You can use filters with custom monitoring profiles to reduce the n Per device component filtering is viewed in the data aggregator UI. The device's filter report lists only the components passing the filter for each monitoring profile slash metric family pair, w You can sort a list of non-filter versus filter components consecutively. The network interface monitoring profile is the only out-of-the-box monitoring profile containing the interface metric family. In addition to component filtering, metric filtering can also be used to reduce the database requirements of collected data. Metric filtering reduces the database requirements of collected data by reducing w Metric filtering is administered on a per metric family basis with a custom monitoring profile. Metric filtering applies only to metrics collected through SNMP. For more detailed information, click the information bubble in the top right corner to load the product page. From there, you can go to the product doc