Skip to content

efcaguab/tracking

Repository files navigation

Peskas: tracking domain

Peskas is an advanced intelligence platform for artisanal fisheries. One of the key features of Peskas is the ability to integrate spatial data from boat tracks with the catch data. From the data architecture point of view, Peskas is divided into “data domains”, each domain plays a bounded role and should function largely independently from other domains. The spatial tracking data is one of Peskas domains and this repository contains all code for its internal data pipeline.

Pipeline description

Pipeline input(s): Tracking data obtained from the Pelagic Data Systems API.

Pipeline product(s): A BigQuery dataset with clean catch data. The dataset includes tables with (i) the full clean dataset.

Stages:

  1. Input data is retrieved on a daily basis at midnight (UTM) by a Google Build Service from the Pelagic Data Systems restful API. Daily tracks are stored in a Google Storage bucket. In addition, the storage bucket is checked every time in case missing days exist.
  2. Changes in the storage bucket that match the daily update prefix are automatically detected and trigger a Google pub/sub notification.
  3. The pub/sub message triggers a call to an API processing function. This API reads the data from the storage and aggregates the data. This data is then deposited in a Google BigQuery table. The aggregation function runs serverless in Google CloudRun.

Deployment steps

Authentication

Ingestion

Aggregation

Continuous development

Scheduling

Support & Contributing

For general questions about the Peskas Platform, contact Alex Tilley. For questions about Peskas’ code and technical infrastructure, contact Fernando Cagua.

Authors

Fernando Cagua

Alex Tilley

License

The code is available under the MIT license.

About

Tracking data domain at Peskas

Topics

Resources

License

Stars

0 stars

Watchers

1 watching

Forks

Packages

 
 
 

Contributors