CALL FOR PAPERS
1st Workshop on High Performance Serverless Computing
Serverless computing represents the next significant leap forward with respect to abstracting commoditized computing resources. The main principle of serverless computing is that compute providers deploy and manage the entire compute environment---from hardware through to runtime in which an application executes; users, therefore, think entirely at the level of their application without regard for the underlying computing environment. For example, the function as a service (FaaS) model allows users to first register a programming function (e.g., in Python) and then they may invoke that function many times with different input arguments.
Serverless computing is poised to become not only the face of cloud computing in the commercial world but also a model for remote and distributed computing more broadly. The 1st workshop on high Performance Serverless Computing will provide the high performance and distributed computing community with a dedicated forum for discussing foundational research in serverless computing in both industry and academia. The workshop will bring together these communities and provide a forum for innovation in both the high performance and distributed systems that underlie serverless computing platforms and also the use of serverless models for abstracting traditional high performance and distributed computing systems.
TOPICS OF INTEREST
The focus of the workshop includes, but is not limited to, the following topics:
- Developing high performance serverless computing platforms
- Federation of serverless computing platforms
- Container management and scheduling
- Programming paradigms for serverless computing
- Data management for serverless computing
- Applications of serverless computing
- Benchmarking and performance of serverless computing
- On-demand and event-based computation
- Serverless economics and billing models
- Fault tolerance and reliability
- Data-intensive workloads and tools
- Productivity tools (Debugging, Profiling, Logging, etc)
- Stream and events processing
- Security in serverless
- Heterogeneous computing (GPU, Accelerators, exotic architectures)
SCHEDULE AND SUBMISSION PROCEDURE
- Papers due:
April 9, 2021 AoEApril 16, 2021 AoE (Hard deadline)
- Notifications: April 30, 2021
- Camera ready: Early May, 2021
- Workshop date: June 25, 2021
Authors are invited to submit papers describing unpublished, original research. All submitted manuscripts should be formatted using the ACM Master Template with sigconf format (please be sure to use the current version). All necessary documentation can be found at: https://www.acm.org/publications/proceedings-template.
Papers may be either full (8 pages) or short (4 pages) including all text, figures, and references. Papers will be peer-reviewed (single blind) , and accepted papers will be published in the workshop proceedings as part of the ACM Digital Library.
Papers conforming to these guidelines should be submitted through EasyChair: https://easychair.org/my/conference?conf=serverless2021
- Yadu Babuji, University of Chicago and Argonne National Laboratory
- Kyle Chard, University of Chicago and Argonne National Laboratory
- Ian Foster, University of Chicago and Argonne National Laboratory