Keynote Speakers
Ahmed E. Hassan, Queen’s University, Canada
Talk title: Software Performance Engineering for Foundation Model-Powered Software (FMware)
Foundation Models (FMs), such as Large Language Models (LLMs), are transforming the landscape of software development, unlocking unprecedented opportunities. While these models demonstrate remarkable potential, evolving FMware into production-ready solutions requires sophisticated engineering across multiple domains. A pivotal and often under appreciated aspect is performance engineering, which ensures FMware achieves critical performance goals like high throughput and low latency—key factors for user satisfaction and financial success. Addressing performance considerations early in development can avoid the significant costs and inefficiencies of post-deployment optimization. Given FMware’s substantial computational demands, efficient hardware utilization becomes paramount. Continuous performance engineering plays a vital role in sustaining optimal performance over time.
In this talk, I will emphasize the importance of Software Performance Engineering (SPE) for FMware, focusing on four major challenges: designing cognitive architectures, establishing robust communication protocols, fine-tuning and optimizing performance, and efficient deployment. These insights are drawn from comprehensive literature reviews and hands-on experience in developing advanced FMware solutions. Join me as we explore these challenges, current best practices, and innovative strategies for the software engineering community
Short Bio
Ahmed E. Hassan is a Mustafa Prize Laureate – commonly equated to a Nobel-level recognition – and a Fellow of ACM, IEEE, and AAIA, as well as an NSERC Steacie Fellow, Canada’s most prestigious mid-career research award across all fields of science and engineering. He holds the Canada Research Chair and the NSERC/BlackBerry Industrial Research Chair in Software Engineering at Queen’s University and is among the world’s most cited Software Engineering (SE) researchers. He is the only individual to receive both the ACM SIGSOFT Influential Educator Award (2019) and the IEEE TCSE Distinguished Educator Award (2020), the highest honors for SE educators from the world’s two largest professional societies. As the founder of the AI-Augmented SE, MSR and AIware communities and a member of the Royal Society of Canada, his career spans over three decades, including leadership roles in both
industrial research (e.g., IBM Almaden, BlackBerry) and academia.
Marc Brooker, AWS, USA.
Talk title: Great Performance for Bad Days
Most traditional approaches to performance measurement and optimization focus on performance under good conditions. Performance during bad times (during and after overload, during and after failures, sudden workload changes, etc) is equally important to customers and operators of systems at all sizes. In this talk, I’ll look at what it takes to keep performance high during adverse conditions, including avoiding and reacting to metastable failure modes.
Short Bio
Marc Brooker is a Distinguished Engineer at Amazon Web Services, where he focusses on AI inference, databases, and serverless. He’s particularly interested in building and operating large-scale systems, formal methods, system performance, and optimization. During his 16 years at AWS, Marc has worked on the teams behind Aurora, EC2, Lambda, EBS, Bedrock, and multiple other AWS products. He holds a PhD in Electrical Engineering from the University of Cape Town, South Africa.