This Dagstuhl Seminar will convene researchers from diverse intellectual communities within computer science who share a common interest in software performance engineering (SPE) : making software run fast or otherwise consume few resources such as time, storage, energy, network bandwidth, etc. With the end of Moore’s Law and Dennard Scaling, microelectronic miniaturization is no longer increasing application performance. SPE has emerged as a promising and widely accessible way to improve performance, gaining increased attention across multiple research areas and industries. SPE encompasses the many ways of developing fast code for a wide range of applications and studying general methodologies and tools to improve and measure performance, including tailoring software to hardware features, algorithms, compiler technology, etc.

Despite the arguably growing importance of SPE, research in the field is scattered across the traditional areas of computer science. One of the goals of this Dagstuhl Seminar is to coalesce a community of researchers to advance SPE as a rigorous and principled scientific subfield of computer science in its own right, comparable to operating systems, algorithms, compilers, machine learning, etc.

In this seminar, we will invite participants who represent diverse areas embodying different expertise, visions, and research foci related to software performance. During the seminar, they will engage in open discussions to explore common interests, methodologies, concepts, and goals across fields. The discovery of new connections among participants will enable lessons from one field to resonate with others and lead to a broader and more systematic body of knowledge for SPE overall.

Creative Commons BY 4.0

Chen Ding, Martin Farach-Colton, Charles E. Leiserson, Peter Sanders, and Yihan Sun