Kinetica is an analytics platform and database designed to handle extremely large and complex datasets with ease. It is distributed, in-memory, and GPU-accelerated.
Combined into a single platform are all the facilities needed to process real-time streaming data, geospatial data, graph relationships, and machine learning. Kinetica can be accessed via SQL, fast NoSQL-style key-value lookups, and REST, along with a variety of language bindings. It includes a complete set of GUI administration tools and a complete dashboard tool that features GPU-accelerated geospatial and graph rendering.
This overview will get you acquainted with the platform before you install your own Kinetica instance or cluster. The goal is to get you up and running quickly so that you can begin testing Kinetica with your own data.
Following this section, there are six additional tutorials that are designed to show off the main features of Kinetica:
- Download and Install Kinetica – A full walkthrough of installing Kinetica (either a single node instance for quick testing or a cluster for performance at scale)
- Importing Data – An overview of a few different methods for getting your data into Kinetica quickly while also giving you a few datasets to import and test out
- Querying with SQL – We give you some queries to run against the imported demo datasets that highlight the breadth of Kinetica’s SQL compatibility and capability
- Location Analytics – We give you additional queries that instead focus on some of Kinetica’s geospatial analytics capabilities
- Reveal Dashboards – We walk you through creating an insightful Reveal dashboard for the demo datasets that highlights some of the data analytics and manipulation available to you
- ML-Powered Analytics – A brief look into one of Kinetica’s defining and innovative features — the Active Analytics Workbench — that will get you leveraging Machine Learning to glean useful predictions from the demo datasets
It’s important to provide an overview of the Kinetica platform, so you’re familiar with some names and terms unique to Kinetica. There are four main tools that help manage or interact with Kinetica:
KAgent – Cluster Management and Maintenance
KAgent is the Kinetica installation and cluster management tool. From KAgent, you can create new clusters, scale your existing clusters, review cluster details, update the security configuration, create a backup or set up a schedule, manage Docker registries for Kubernetes and AAW, and start, stop, or restart any Kinetica service.
GAdmin – Database Administration
GAdmin is the administrative interface of the Kinetica database. It is a single stop for the configuration, security, logging, and monitoring of your Kinetica database. GAdmin also provides the means for data ingestion, querying via SQL and REST APIs, executing user-defined functions (similar to stored procedures), and monitoring long-running jobs. Kinetica has access to five demo datasets containing a variety of data types and sharding strategies that are perfect for testing purposes; these datasets are available via the GAdmin Demo page.
Reveal – BI and Analytics Dashboard
Bundled with Kinetica is Reveal, a web-based BI visualization framework for querying and charting of data in an easy, interactive way. Reveal has a rich set of tools for making dashboards with charts, diagrams, and map visualizations. Multiple users can access these, and dashboards can be shared with others.
AAW – Advanced Analytics Workbench
AAW simplifies and accelerates data science and machine learning workflows in a scalable fashion. With AAW, users can create and manage batch or continuous ingests; create and manage TensorFlow featuresets and datasets; create, manage, and deploy RAPIDS and black box models and analytic functions; make inferences on models; and even audit models.