Data Development Cycle

We are planning that each group compiling state data will be operating on an approximately one year data development cycle. Because of the number of partners who will be submitting data, we need to stagger data delivery so that the end of each cycle is not an impossible-to-handle data deluge.

Figure 1 below displays the data development cycle. After submission, review, and acceptance of the Statement of Work for the cycle, each participant will prepare a small test data set to prototype the data delivery content and format for each deliverable product. The NGDS development team at AZGS, in collaboration with developers at our regional training and server hubs, will review this prototype and test integration of the data with the NGDS system. Debug iterations will be made between the NGDS system and each data producer until the prototype is demonstrated to work and provide the necessary content. At this point the project management team will give the data provider a go-ahead for development of the full data product for that cycle. The prototype dataset should at that point be made ‘live’ (accessible online) in the system, but flagged as a development data set. It is strongly recommended that as data are developed, they are incrementally added to the prototype delivery dataset so that content is available as early as possible, and problems with scaling will become apparent if they emerge. When the data producer has completed the data development, they will make a final data submission for technical review and approval. Ideally the final submission will simply be a declaration that the live dataset evolved from the prototype is complete and ready for final review. Upon approval by the project management, the data producer may submit their SOW for the next data development cycle.