Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for high-throughput parallel data transmission. In some implementations, a data connector module is configured to provide an application server with access to a data set of unstructured data. One or more computers receive a first request associated with a first application or service, and data identifying a second application or service different from the first application or service. A plurality of second requests are generated and distributed to processing nodes to generate results for the second requests. Result data derived from the results for the second requests to the second application or service.