May 6 , 2025 - Toronto, Canada
8th Workshop on Hot Topics in Cloud Computing Performance (HotCloudPerf 2025)
at ICPE 2025
Overview
The HotCloudPerf workshop proposes a meeting venue for academics and practitioners, from experts to trainees, in the field of cloud computing performance. The new understanding of cloud computing covers the full computational continuum from data centers to edge resources to IoT sensors and devices. The workshop aims to engage this community and to lead to the development of new methodological aspects for gaining a deeper understanding not only of cloud performance, but also of cloud operation and behavior, through diverse quantitative evaluation tools, including benchmarks, metrics, and workload generators. The workshop focuses on novel cloud properties such as elasticity, performance isolation, dependability, and other non-functional system properties, in addition to classical performance-related metrics such as response time, throughput, scalability, and efficiency.
Acknowledgement
The HotCloudPerf workshop is technically sponsored by the Standard Performance Evaluation Corporation (SPEC)’s Research Group (RG) and is organized annually by the RG Cloud Group. HotCloudPerf has emerged from the series of yearly meetings organized by the RG Cloud Group, since 2013. The RG Cloud Group group is taking a broad approach, relevant for both academia and industry, to cloud benchmarking, quantitative evaluation, and experimental analysis.
Program (preliminary)
09:00 Opening
Session 1: AI performance
Chaired by Klervie Toczé
09:10 Keynote: “The Cloud AI Imperative: Performance, Regulation, and Intelligent Automation” by Padma Apparao
10:10 SPEC RG Cloud highlights
10:30 Morning coffee break
Session 2: Resource management and performance evaluation
Chaired by André Bauer
11:00 Sharod Roychoudhury, Dheeraj Chahal, Rekha Singhal
GenAI for Bottleneck Detection in Cloud Architecture
Full Workshop Paper
11:25 Anders Nou, Sacheendra Talluri, Alexandru Iosup, Daniele Bonetta
Investigating Performance Overhead of Distributed Tracing in Microservices and Serverless Systems
Short Workshop Paper
11:40 Zhiqi Li, Ruiqi Yu, Jianshu Liu
IrisBench: An Open-source Benchmark Suite for Video Processing Systems in Cloud
Full Workshop Paper
12:15 Klervie Toczé
Introducing Resource Awareness Levels in Edge Computing Resource Management
Short Workshop Paper
12:30 Lunch break
Session 3: Memory and service performance
Chaired by Klervie Toczé
14:00 James McMahon, Vinita Pawar, Ryan Stutsman
Remote Memory Prefetching: Is Coarse-grained Fine?
Full Workshop Paper
14:25 Vinita Pawar, Ankit Bhardwaj, Ryan Stutsman
ObjectTier: Non-invasively Boosting Memory Tiering Performance
Full Workshop Paper
14:50 Sándor Battaglini-Fischer, Bálint László Szarvas, Nishanthi Srinivasan, Xiaoyu Chu, Alexandru Iosup
FAILS: A Framework for Automated Collection and Analysis of LLM Service Incidents
Full Workshop Paper
15:15 Afternoon coffee break
Session 4: Applications & performance modeling
Chaired by TBA
16:00 Yannik Lubas, Martin Straesser, Ivo Rohwer, Samuel Kounev, André Bauer
Microservice Applications and Their Workloads on Github
Full Workshop Paper
16:25 Panel: Performance modeling for the computing continuum
17:20 Closing
17:30 End of HotCloudPerf 2025
Keynote
Padma Apparao: The Cloud AI Imperative: Performance, Regulation, and Intelligent Automation
Principal Engineer, Intel
Description: Cloud-based AI is driving innovation, but optimizing its performance presents significant hurdles. Data bottlenecks, diverse hardware, fluctuating resource needs, and evolving AI regulations all challenge businesses. This talk explores how the cloud is becoming essential for AI deployment, especially for organizations lacking in-house infrastructure. The discussion will highlight how AI agents are enabling intelligent monitoring and management of cloud AI workloads, leading to proactive issue resolution and enhanced efficiency. Strategies for overcoming these challenges, including automated tuning, hardware optimization, and dynamic resource allocation, will be presented. The impact of data location, hardware acceleration, and network latency on AI applications will be examined, emphasizing the need for solid performance evaluation. Finally, the critical role of data governance and AI regulations will be addressed. As privacy and ethics gain importance, compliance becomes paramount. The talk will demonstrate how intelligent resource scheduling, model compression, and hyperparameter tuning can enhance scalability, efficiency, and regulatory adherence, enabling responsible and cost-effective cloud AI.
Bio: Padma Apparao is a Principal Engineer and AI and Cloud Performance Architect at Intel. She began her career at Intel as a server performance analyst in 1996, after receiving a Ph.D. in Computer Science from the University of Florida. Since then, Padma has worked across multiple market segments, optimizing workloads and benchmarks for servers, clients, and desktops.
Currently, Padma is the AI architect in the Intel Federal Solutions group, providing customized holistic AI solutions to US federal agencies. Her current focus is on RAG-based AI solutions, working on the holistic performance evaluation of RAG and compliance methodologies. Padma is well recognized within Intel for her performance methodology skills and leads the Open-source OPEA (Open Platform for Enterprise AI) Evaluation Working Group across the industry.
Her expertise encompasses empirical performance studies in AI and cloud computing, including observation, measurement, and surveys, as well as performance analysis using the latest and emerging technologies. Padma specializes in end-to-end performance engineering for pipelines and workflows in cloud environments, developing and utilizing tools for monitoring and studying cloud performance.
She regularly writes AI newsletters that reach over 10,000 people at Intel. Padma has several published papers, has delivered many invited talks, participated in AI panels, and holds 23 patents. In her spare time, she enjoys solving crossword puzzles.
Topics
Empirical performance studies in cloud computing environments, applications, and systems, including observation, measurement, and surveys.
Performance analysis using modeling and queueing theory for cloud environments, applications, and systems.
Simulation-based studies for all aspects of cloud computing performance.
Operational techniques for self-organization, resource management, and scheduling in cloud environments, e.g. service meshes, auto-scaling, auto-tiering.
End-to-end performance engineering for pipelines and workflows in cloud environments, or of applications with non-trivial SLAs.
Tools for monitoring and studying cloud computing performance.
General and specific methods and methodologies for understanding and engineering cloud performance.
Methodological and practical aspects of software engineering, performance engineering, and computer systems related to hot topics in cloud performance, e.g. serverless, microservices, non Von Neumann architectures, virtualization/containerization.
Case studies on cloud performance and its interaction with the computing continuum, including benchmarking, exploratory studies, dataset collection and negative results.
Sustainability and energy-efficiency in cloud computing environments, applications, and systems.
Network, storage and accelerators in the computing continuum.
Cloud computing environments, applications, and systems should be understood in the broad sense and include works looking at the computing continuum (i.e. IoT-edge/fog-cloud).
Important Dates
January 17, 2025, AoE February 1, 2025, AoE
January 24, 2025, AoE February 1, 2025, AoE
February 17, 2025 February 26, 2025, AoE
February 26, 2025 March 19, 2025
May 6, 2025
Abstract due (informative, new submissions welcomed)
Papers due
Author Notification
Camera-ready deadline
Workshop day
Submission Types
Full-papers (6 pages including tables and figures but not references and appendices)
Short-papers (3 pages including tables and figures but not references and appendices)
Talk only (1-2 pages, not included in the proceedings).
Format
The format of the submissions is single-blind and needs to follow the ACM format of the companion conference, ICPE.
All presented papers will have a good amount of time allocated for Q&A plus feedback. In addition, the presentation session will be wrapped up by a 10-15 min discussion.
ACM Instructions for Authors
By submitting your article to an ACM Publication, you are hereby acknowledging that you and your co-authors are subject to all ACM Publications Policies, including ACM’s new Publications Policy on Research Involving Human Participants and Subjects. Alleged violations of this policy or any ACM Publications Policy will be investigated by ACM and may result in a full retraction of your paper, in addition to other potential penalties, as per ACM Publications Policy.
Please ensure that you and your co-authors obtain an ORCID ID, so you can complete the publishing process for your accepted paper. ACM has been involved in ORCID from the start and we have recently made a commitment to collect ORCID IDs from all of our published authors. The collection process has started and will roll out as a requirement throughout 2022. We are committed to improve author discoverability, ensure proper attribution and contribute to ongoing community efforts around name normalization; your ORCID ID will help in these efforts.
Submission Site
Articles and talk-only contributions are required to be submitted via HotCRP.
Call for Papers
You can find the full Call for Papers (CfP) here: CfP
Organizing Committee
Klervie Toczé, VU Amsterdam, the Netherlands
André Bauer, Illinois Institute of Technology, USA
Dragi Kimovski, Klagenfurt University, Austria
Daniele Bonetta, VU Amsterdam, the Netherlands
Advisory Board
Cristina L. Abad, Escuela Superior Politécnica del Litoral, Ecuador
Nikolas Herbst, University of Würzburg, Germany
Alexandru Iosup, VU Amsterdam, the Netherlands
Program Committee
Cristina Abad, Escuela Superior Politecnica del Litoral
Auday Al-Dulaimy, Mälardalen University
Atakan Aral, University of Vienna
Matt Baughmann, University of Chicago
Andre Bondi, Software Performance and Scalability Consulting LLC
Lubomir Bulej, Charles University
Tommaso Cucinotta , Scuola Universitaria Superiore Pisa
Tiziano De Matteis, Vrije Universiteit Amsterdam
Ian di Dio Lavore, Politecnico di Milano
Antonino Galletta, University of Messina
Maxime Gonthier, University of Chicago
Wilhelm Hasselbring, University of Kiel
Nikolas Herbst, University of Würzburg
Dragi Kimovski, University of Klagenfurt
Tania Lorido, Roblox
Narges Mehran, University of Salzburg
Zahra Najafabadi, University of Innsbruck
Prateek Sharma, Indiana University Bloomington
Sacheendra Talluri, Vrije Universiteit Amsterdam
Klervie Toczé, Vrije Universiteit Amsterdam
Petr Tůma, Charles University
Chen Wang, IBM