REAL-TIME, EDM-INSPIRED SONIFICATION OF THE ACTIVITY OF A SUPERCOMPUTER
2026
Read REAL-TIME, EDM-INSPIRED SONIFICATION OF THE ACTIVITY OF A SUPERCOMPUTER (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-21874.txt.
Introduction
Marco Alunno and Paolo Bientinesi explore real-time, EDM-style sonification of supercomputer activity, turning node data into live tracking of system behavior. Because such monitoring never ends, the sonification must stay clear and lively over long spans. Rather than imposing a preset musical style, the authors sought one the data could support, settling on EDM, whose structure and timing suit continuous, data-driven processes and long listening. The work combines monitoring over debugging, real-time over post-mortem interpretation, and endless, coherent-in-style music generation.
Catalog record
| YEAR | 2026 |
|---|---|
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35755 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.21874
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
REAL-TIME, EDM-INSPIRED SONIFICATION OF THE ACTIVITY OF A SUPERCOMPUTER Abstract The project described in this paper explores the informative sonification of data received in real time from a supercomputer. These data capture the current activities in all the nodes of the computer, therefore, their sonification functions as a form of continuous monitoring of the nodes’ behavior and, by extension, of the system as a whole. Because such monitoring is theoretically unending, the resulting sonification must be musically capable of conveying information through sound in a way that remains both intelligible and engaging over long durations. Rather than imposing a predefined musical style onto the data, we sought to identify one which the data themselves could plausibly support. From a small set of candidates, we selected EDM because it is a family of genres whose structural and temporal characteristics align well with continuous, data-driven processes and long-term listening. Through this style-based approach, this research builds on the long tradition of computer data sonification while uniquely combining three elements rarely addressed together: monitoring (rather than debugging) as the primary goal, real-time (rather than post-mortem) data interpretation, and generation of virtually infinite and stylistically coherent (rather than incongruous) music. 1 Introduction This project is driven by a conceptual metaphor: A supercomputer, with its thousands of cores relentlessly executing computational tasks and perfectly coordinated by a central management system, evokes the image of an immense mechanized orchestra under the control of a conductor’s baton. Further, a supercomputer is organized according to a hierarchy in which the individual cores are part of nodes, which, depending on their typology, are in turn grouped into partitions. Within the metaphor, if a supercomputer is an orchestra, the partitions are the orchestra’s instrumental families, and the nodes are the players whose activity is controlled by the management system (conductor). Since the nodes’ activity is summarized through a set of features expressed as numerical values, converting these values into sound would produce an acoustic image of the behavior of each node and, consequently, a radiograph of the computer as a whole. The goal of this work is to design such a conversion and to compose an informative sonification for the acoustic monitoring of a supercomputer that is in real time, uninterrupted, and never-ending. Monitoring refers to the continuous, low-overhead observation of a running system aimed at understanding its state, performance, and long-term trends. Unlike debugging, monitoring does not involve the investigation, identification, or correction of abnormal behaviors. In our case, the monitored events are the activities and processes of a supercomputer, which are traditionally presented through data visualization techniques [ 26 ] . Because monitoring may extend over long periods of time, offloading the representation of data from the visual to the auditory domain can be advantageous: as noted in prior works [ 31 , 29 , 5 ] , a sonification reduces the need for undivided visual attention and allows the listener to remain engaged without continuously focusing on a screen. As target listeners, we picture supercomputer users and administrators, as well as ordinary passersby in spaces around the supercomputer facilities. The supercomputer we chose to monitor, Kebnekaise, is operated by the High-Performance Computing Center North (HPC2N) at Umeå University, Sweden. At the time of writing, Kebnekaise comprises 10 partitions, 95 nodes (for a total of 206 CPUs and 74 GPUs), whose activities are overseen by the Simple Linux Utility for Resource Management (Slurm) workload manager. Slurm continuously monitors the state of the processors of each node, and schedules incoming requests to available resources (Sec. 3 ). Additionally, Slurm has also the ability to collect various data, and to produce informative statistics about usage, conditions, and activities of the nodes. In our study, we focus on three metrics—namely, the number of running processes, the amount of physical memory used, and the amount of input/output activity. In detail, information about these metrics is periodically extracted from each node, gathered into three arrays in a local server, and then reorganized as three streams that are sent to the internet. The three streams are received by a different computer where a Python script reads and forwards them, via OSC, to the sonification code in SuperCollider 1 1 1 The sonification code in SuperCollider is available at https://github.com/pupil72/kebne-sonification . A central feature of the project, that enables users to gain an immediate sense of the levels of activity within Kebnekaise, is that the sonification refreshes in real time with the incoming data stream. Moreover, the regular, periodic renewal of the sonification leverages anticipation as a perceptual mechanism, helping to sustain the listener’s engagement during continuous monitoring [ 14 ] . Sustained engagement is particularly important in this project, because the duration of the sonification is undefined, lasting at least until the supercomputer is restarted, which normally happens every few weeks or months. However, engagement is not a direct consequence of the anticipation effect. For anticipation to be operative, it must allow the listener to form predictions—however correct or incorrect—about forthcoming sonic events. Consequently, for sound to be engaging, it must be organized in a way that encourages predictive listening. When the listener possesses prior knowledge of that organization, engagement may arise immediately; otherwise, the principles governing the organization may gradually emerge through extended exposure [ 13 ] . Our search of suitable forms of organization ultimately led us to conceive sound in terms of musical style—that is, sound structured according to a set of rules, possibly known by most, that constrain and define its behavior. The preconditions of regular changes and indefinite length suggested us to use a repetitive, non-teleological musical style. Out of a shortlist of candidates, among which drone and ambient music, we soon identified EDM as the most appropriate, since it subsumes an umbrella of genres whose formal structure and temporal behavior are well suited to continuous, data-driven processes and extended listening. In fact, many real-world applications of EDM already support the practice of extended listening, as this style of music is primarily intended for use in dance-oriented settings that can last from several hours to an entire day or even longer. In what follows, we first provide an overview of previous research on the sonification of supercomputing. We then briefly describe several aspects related to the mechanisms involved in handling and managing a supercomputer’s resources. The next two sections discuss, respectively, the questions we addressed in order to advance the project and the solutions we ultimately adopted, together with the technical characteristics of the sonification. We conclude by outlining potential improvements and future extensions of the project. 2 Literature review The development of sonification for computer and supercomputer processes has transitioned from the accidental detection of electromagnetic interference to the intentional design of musical frameworks for continuous system monitoring. In the mid-twentieth century, early engineers and programmers relied on unintended audio cues (e.g. tuning AM radios to pick up CPU interference or attaching speakers to index registers) to discern the state of running compilers [ 7 , 28 ] . These primitive methods established the fundamental concept that computer activity possesses a distinct sound life that can be meaningfully interpreted by the human ear. However, while these early instances provided a form of monitoring, the field quickly moved toward