I'm seeking some advice as to how to create a sustainable and reliable structure for manituplating data from external sources (i.e. outside of the internal Qlikview server setup). Basically my situation is that we have a number of datasources (primarily SQL databases) across multiple sites in Europe and require data to be moved over the network to the central Qlikview servers where Publisher, Access Point etc reside. We have been dealing with this on a small scale to date with a couple of sites having Qlikview PE installed and a windows task scheduled to run the Qlikview application which in turn executes the SQL query and then copies the qvd from the external site to the internal Qlikview servers.
However this is now being rolled out to new sites and there is a new demand for data to be extracted on an hourly basis and then incrementally consolidated with data from other sites. We are limited in the sense that the data can only go over the network as a qvd (i.e. not the raw SQL query) and due to bandwith issues the file size has to be as small as possible. My main concern is that what happens if one or multiple servers crash and this affects the SQL queries which essentially will corrupt the incremental load process.
Has anyone got any practical advice or best practice suggestions for this?