All times in Anywhere on Earth (AoE) timezone.
Welcome message from the Chairs.
17.50-18.10: Energy-Efficient Resource Management for Real-Time Applications in FaaS Edge Computing Platforms. Shahrokh Vahabi (University of Pisa); Francesca Righetti (University of Pisa); Carlo Vallati (University of Pisa); Nicola Tonellotto (University of Pisa)
18.10-18.30: Scheduling Energy-Aware Multi-Function Serverless Workloads in OpenFaaS Karim Djemame (University of Leeds); Raulian Chiorescu (University of Leeds)
Serverless computing has gained enormous popularity in recent years, promising seamless scalability, fine-grained pricing, and simplified management. According to the serverless paradigm, providers should take care of most operational issues, with users only responsible for deploying applications and specifying desired Quality of Service (QoS) levels. This fostered the development of the so-called Function-as-a-Service (FaaS) computing paradigm, that is nowadays offered by most popular cloud providers.
However, current implementations of serverless struggle to fully realize this computing paradigm vision, and serverless adoption for many applications is slowed down. For instance, latency-sensitive applications cannot tolerate unpredictable cold start delays of FaaS platforms; also, stateful serverless applications are currently not well supported.
These challenges become even more difficult to solve as we consider the adoption of serverless at the edge of the network as well as in the emerging cloud-edge continuum, where resource constraints and their heterogeneity represent a serious challenge for QoS guarantees.
In this workshop, we solicit high quality contributions that fit with the overarching theme of guaranteeing QoS levels in serverless computing systems and applications.
We invite submissions of original research papers, as well as vision papers and experience reports.
The workshop aims to promote discussions on architectures, mechanisms, and algorithms to support QoS-aware deployment and execution of serverless applications. Topics of interest include (but are not limited to):
Authors are invited to submit papers electronically through CMT.
Submitted manuscripts must represent original unpublished research that is not currently under review for any other conference or journal. Manuscripts are submitted in PDF format and may not exceed six (6) pages, including figures, tables, and references, in the standard ACM format for conference proceedings. All manuscripts will be reviewed and judged on correctness, originality,technical strength, rigour in analysis, quality of results, quality of presentation, and interest and relevance to the conference attendees. Your submission is subject to a determination that you are not under any sanctions by ACM.
The workshop adopts a double-blind review process, where neither authors nor reviewers know each other’s identities. This means that papers must not list or otherwise identify the authors. References to previous work should be done in the third person so as to not reveal the identities of the authors. Submissions that are not anonymous may be rejected without review.
At least one author of each accepted paper must be registered for the conference (Note: full registration, non-student, non-workshop) in order for the paper to be published in the proceedings.
Authors: At least one author of each paper must register for the conference. A full registration (non-student, non-workshop) is required in order for the paper to be published in the proceedings.
Participants: All workshop participants must register under the UCC 2023 workshop day registration fee or the conference fee.
Please refer to the UCC website for more information about the registration process.