The award winning Axel system, platform, and components
Aha! uses a proven platform architecture whose core features are implemented in Java, with web services through JBoss for true multi-tenancy in logic, data, and security, and encrypted security tokens for each call, XML-based parameters including facilities mashups, SSL communications, distributed logging for integration, MySQL databases, a data services SOA for external data and interface integration, nested and federated ETL supporting 500 adapters including five Aha!-specific components, and secure transports for publish and subscribe and SFTP.
Axel Data Services leverages high-performance redundant data capture using HDFS (Hadoop file system) clusters for real-time data capture and aggregation of web experiential data, scalable high-performance capture/analysis using Apache’s Hive and Hadoop map/reduce framework, and commercially available cloud frameworks to provide scalable analysis services across single and multiple simultaneous client implementations.
The Aha! development process is strictly driven by productization and release objectives. The release process is driven by the product roadmap, which drives specifications, development, test, deploy, and operational stages. The development environment is based on Eclipse IDE (Java, C, C++), Subversion software version control, Mantis issue tracking, and Xdoclet for web services wrapping. Development is on Windows-based systems with nightly build of a deployable system concurrent with unit test runs and reporting. Integration testing is isolated to integration testing-only servers. QA Testing is isolated to QA testing servers only. Deployment scripts are developed and test deployments run on a shadow environment. There are two production target system environments, one for demos and one for production.
Axel data and modeling flow
The Axel core receives business data from a variety of disparate sources: customer service and billing, CRM data, sales and marketing campaign data, web analytics, and even social networking data. This real-time or near-real-time data is fed to Axel continuously, where it undergoes scoring, regression modeling, and simulation and optimization modeling. KPIs are generated as defined by the company itself, and mathematical relationships among KPIs are maintained in the Axel core.
