Study Of Parallel Production Technology Of Re-mote Sensing Product And Its Application

Remote Sensing Applications product is a kind of data processing that is data-intensiveand Compute-intensive. As people’s demand for timeliness of information products continu-esly improve, how to accelerate the production of remote sensing applications, complete theinformation products in the shortest possible time, has become a technical problem to besolved.This paper analyzes the domestic and foreign research status of remote sensing applica-tions products first, then focuses on the parallel-driven technology of remote sensing produc-tion and its applications in task processing subsystem of remote sensing production system.The main contents of this paper include the following aspects:1Design and implement the production subsystem of remote sensing products system.Based on fully understanding and analyzing user needs, complete of the design of the systemarchitecture, implement the production of topic remote sensing products, algorithm driversand other core functions of the business system.2Interface specification of remote sensing production system and production algorithms.There are many information products product algorithms realized by many different researchgroups. In order to en-sure all the algorithms integrated into information products productionsystem can operate normally, the remote sensing production system can call algorithms tocomplete the information production, we need to develop a uniform interface specification,the purpose is to unify the development runtime environment of all product algorithms, stipu-late the interface and interactive way between the remote sensing production system and theproduct algorithms. 3Design parallel modes of remote sensing products production. There are two levels ofparallelism, one task-level parallelism in the production, namely the production of multipletasks in parallel; Second component-level parallel, when there are two or more components ofan algorithm can be produced during production, driving several components at the sametime.4Remote sensing products production based on directed acyclic graph. Designed modelof directed acyclic graph process and discusses the construction, storage, parallel drive andother related issues. Using a directed acyclic graph to describe the dependencies among thecomponents of the algorithms, fully satisfy the description need of dependencies among re-mote sensing products. Using cross-linked diagram to complete the description to produceproduct in actual.