Sonifying Performance Data to Facilitate Tuning of Complex Systems: Performance Tuning: Music to My Ears

摘要

In the modern computing landscape, the challenge of tuning software systems is exacerbated by the necessity to accommodate multiple divergent execution environments and stakeholders. Achieving optimal performance requires a different configuration for every combination of hardware setups and business requirements. In addition, the state of the art in system tuning can involve complex statistical models, which require deep expertise not commonly possessed by the average software developer. This paper presents a novel approach to tuning complex software systems by leveraging sound to convey performance information during execution. We conducted a scientific survey to determine which sound characteristics (e.g., loudness, panning, pitch, tempo, etc.) are most accurate to express information to the average programmer. As determined by the survey, the characteristics that scored the highest across all the participants were used to create a proof-of-concept demonstration. The demonstration showed that a programmer who is not an expert in either software tuning or enterprise computing can configure the parameters of a real world enterprise application server, so that its resulting performance surpasses that exhibited under the standard configuration. Our results indicate that sound-based tuning approaches can provide valuable solutions to the challenges of configuring complex computer systems.