-
Title
Data Collection Overview -
Description
Desktop Authority is designed to collect essential User Activity, Hardware and Software Inventory into a central SQL database. Data is collected from all clients that execute Desktop Authority during the logon and logoff process.
-
Resolution
The data collection process uses standards-based XML files to store client data. Data is collected from each client and is temporarily stored on the client. Using a push technology, the data is then stored in network cache directories. The network cache directories are managed by the Desktop Authority network services component called the SLServer service. The network services are typically deployed per site in a multi-site network. Data is stored on the network cache to ensure system scalability and reliable data captures, especially during periods of WAN downtime between sites.
From the network cache directories, the data is pulled into the ETL repository on the Operations Master server. This data pull ensures that data is collected from site positioned servers in a scalable and secure fashion to support throttling and load balancing on the database server and network.
The following graphic depicts the flow of the data collection process.
During the client’s session on the network User Activity (logon/logoff/lock/unlock), Hardware/Software Inventory and Patch and Anti-Spyware data are collected and temporarily stored on the client.
Using Named Pipes, the XML data is pushed to a reporting cache. This cache is a hidden share that is created by the SLServer service. Each server that hosts the SLServer service has its own cache. There may be one or more cache shares. The default share location is %programfiles%\ScriptLogic\ETL Cache. This location can be configured in the Service Configuration dialog in the DA Manager. The specified location must exist on the server running the service.
Data is pulled from the server’s caches by the ETLProcessor service and put into the ETL Repository. The data is pulled by one or more file copy threads on a scheduled and throttled basis. The timing for this process can be configured for the ETLProcessor service on the OpsMaster tab of Service Manager in the DA Manager. The default schedule is every 60 minutes.
Once the data is pulled into the ETL repository, it must be transformed and loaded into the Reporting database. When data is found in the ETL repository, the ETLProcessor will copy the files and then transform the data against a set of business rules. Once this is complete it is loaded into the database.