AWIPS II: Improvements in System Stability and Performance

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner
Wednesday, 5 February 2014
Hall C3 (The Georgia World Congress Center )
Michael J. Bettwy, NOAA/NWS, Silver Spring, MD; and S. Garrard, B. Yin, and N. DiPasquale

The Advanced Weather Interactive Processing System (AWIPS) is used by the National Weather Service (NWS) to ingest, analyze and disseminate operational weather data including time-sensitive, high-impact warnings. NWS and Raytheon are working together in the deployment of AWIPS II, the system's next-generation software, which migrates AWIPS I capabilities into a Service Oriented Architecture (SOA). As of August 2013, AWIPS II is in the Operational Test and Evaluation (OT&E) phase at select deployment sites. While new functionality is being delivered to forecasters in the field, system performance and stability also need to remain at acceptable levels so that these advancements can be fully realized.

Components of AWIPS II are distributed and clustered on multiple servers. The Java Virtual Machines (JVMs) comprise the Environmental Data EXchange (EDEX) server and handle ingest, processing, and read/write access to processed data storage, distributed across four primary servers to reduce processing latency. When combined with data services that use advanced serialization techniques, the decision-making visualization is faster. The Eclipse-based Integrated Development Environment (IDE) and GIS interactions for warnings have all contributed to improvements in system stability and performance. This poster presentation summarizes the advancements made in these areas during recent AWIPS II operational builds, as revealed in results from detailed performance test cases, and how these data compare against similar metrics in AWIPS I.

Over the next several years, as more data becomes available and increasingly complex products (like those from GOES-R) come online, AWIPS II will serve as the computational conduit and must remain stable and reliable so that overall operations and collaboration efforts are not negatively impacted. This presentation will also summarize some of these challenges and how they may be mitigated.