File Types and Naming Conventions

Statistical records are forwarded from the device to the SDM server for viewing in a comma-separated value (CSV) file. Before a file is pushed by the device, the device creates a directory by group name for which the statistic belongs (for example, diameter director, system, etc.), if the directory does not exist from a previous push.

Each file is formatted as <UTC timestamp in seconds>.csv (for example, 201112250000.csv).

Each CSV file contains a record for the header containing the statistical attribute name, as well as both the push interval and collection interval records. The first record of each file is a header containing the attribute name. For example, in the “System” directory, a file name of 201112250000.csv can contain the header names of CPU Utilization, Memory Utilization, Health Score, Redundancy State, etc. Also included in the files are both the push interval and collection interval files. For example, if the collection interval is one minute, and the push interval is 15, a collection occurs every minute for 15 minutes. The CSV file in this example contains 15 records.