Velo is a customizable, collaborative, project-centric, knowledge management and analysis framework based on commercial grade products, it provides an integrated environment for secure, collaborative data and knowledge management, analysis, visualization and sharing. It includes a flexible, rich content model which can define any object from file-based contributions to relational content. All content objects (data, knowledge, application, users etc.) are represented as a semantic graph of nodes which connect objects together via a schema of domain-specified relationships, enabling not only to represent and utilize the relationships between data and knowledge objects, but also their link to analytical tools and workflows. Velo comes with several default schemas, including one to support provenance information to enable veracity checks for observational, simulated and analytical results. Importantly Velo provides a tool integration framework which allows deployments to quickly integrate a wide variety of existing analytical tools including workflows, scripts, analysis and visualization tools. Velo has been integrated with other semantic tools such as VisKo (Open Source Visualization Knowledge) in order to provide a more powerful search for tools that can be applied to a given data set. The system also offers transparent remote processing capabilities, allowing the user to utilize remote data, database, HPC and cluster resources, often eliminating the need to move data for further processing. The Velo core system can scale up to any demand by running in a clustered or cloud environment. In addition, Velo can federate data located at a variety of remote locations using its remote nodes, so data does not need to be contained by the Velo repository in order to be utilized by tools. Velo provides secure, role-based access to resources, and it can store classification levels for each resource in support of multi-level security. Velo provides a customizable desktop client as well as a web interface that can be easily tailored to meet the needs of each specific deployment, alternatively a customized client can be developed utilizing the Velo core system API's. Velo is the ideal deployment platform for easy to use analysis and visualization environments, handling all underpinning data and resource needs, in particular for heterogeneous data, distributed resources and high computational demands all characteristics of today’s big data environments. Its success and versatility is demonstrated by its deployments:
Velo has been operationally deployed for use in several US Government agencies, including the Department of Energy's Biological and Environmental Research program, the Department of Homeland Security's Seattle Law Enforcement Gang Task Force, and the Department of Energy's Consortium for Advanced Simulation of Light Water Reactors research. In addition, Velo deployments support the following PNNL Laboratory Directed Research and Development (LDRD) projects: Advanced Simulation Capability for Environmental Management (Groundwater Modeling), Carbon Sequestration Initiative (Geochemistry Modeling), Integrated Regional Earth System Model (Integrated Climate and Energy Modeling), and Chemical Imaging Initiative (Near Real time and integrative analysis of multi-modal experimental data).