The Kiji mission is to help developers use data to create amazing applications.
The traditional data architecture stack is complex and ill-equipped to handle the scale and variety of data now available in today’s digital age. There are new emerging open-source paradigms for data management, such as Hadoop, but there are still significant barriers to adoption, including substantial knowledge of distributed computing architectures and large-scale machine learning techniques.
We see an opportunity to help establish new tools, standards and design patterns for both analytical and operational data processing, specifically with a focus on novel applications of data such as predictive modeling and real-time data applications. These tools should retain best practices from relational data modeling systems, but focus on scalability and fault-tolerance as well as usability to allow for broad adoption across the data ecosystem.
Ultimately, our goals are:
- To improve ease-of-use of Hadoop and HBase
- To accelerate development of data applications
- To maximize performance and stability of HBase for real-time high-availability applications
- To promote broad adoption of Hadoop and HBase as an applications development platform
The Kiji Project uses the ASF 2.0 License.
Community & Partners
The Kiji Project is an open-source community project that seeks contributions from any organization. Please see Contributing to Kiji to learn more about how to contribute.
The project was started by WibiData, a San Francisco-based startup that supports enterprises building real-time, personalized applications at scale, such as product recommendations, mobile offers, and risk assessment. WibiData provides companies training, services, and an SDK for developing data-driven applications. Our framework enables companies to collect and consolidate data about their users, train and deploy predictive models, and scalably serve user-facing applications all from their Hadoop cluster. Find out more about WibiData at wibidata.com.