
۱۳ خرداد - ۱۴ خرداد ۱۴۰۴
Workshop on Job Scheduling Strategies for Parallel Processing




نمای کلی
The 28th Workshop on Job Scheduling Strategies for Parallel Processing (JSSPP 2025) will be held in conjunction with the 39th IEEE International Parallel and Distributed Processing Symposium (IPDPS'2025) in Milan, Italy, from June 3-7, 2025. The workshop is planned as an in-person, two-day event. JSSPP 2025 welcomes novel submissions including Regular Papers, Workload Traces, and descriptions of Open Scheduling Problems.
JSSPP 2025: Call for Submissions
The 28th Workshop on Job Scheduling Strategies for Parallel Processing (JSSPP 2025) welcomes novel, unpublished Regular Papers (RP), descriptions of Open Scheduling Problems (OSP), and Workload Traces (WT) from interesting scheduling domains.
Call for Regular Papers (RP)
JSSPP solicits regular papers that focus on challenges in parallel scheduling, including but not limited to:
- Design of new scheduling approaches.
- Performance evaluation of scheduling approaches, including methodology, benchmarks, and metrics.
- Fulfilling additional constraints in scheduling systems, like job priorities, price, energy requirements (Green Computing), accounting, load estimation, and quality of service guarantees.
- Impact of scheduling strategies on system utilization, application performance, user-friendliness, cost efficiency, and energy efficiency.
- Scaling and architecture of very large scheduling systems.
- Operational challenges: Capacity planning, service level assurance, reliability.
- Interaction between schedulers on different levels (e.g., processor vs cluster) and tenancy domains (single and multi-tenant)
- Interaction between applications/workloads, e.g., efficient batch job and container/Pod/VM co-scheduling within a single system, etc.
- Experience reports from large-scale compute production systems.
- GPU/Accelerator (co)scheduling.
- AI/ML-inspired scheduling approaches.
Call for Workload Traces (WT)
JSSPP welcomes novel, unpublished papers introducing workload traces from workloads of real systems that offer challenges in the context of this workshop. These submissions should include:
- Anonymized (if needed) workload artifacts that describe a significant share of individual units of resource scheduling (jobs, pods, VMs, functions, etc.) during a period of life in a parallel computing system.
- A description of the parallel system running these workloads.
- An analysis of the traces, modeling key workload features and highlighting scheduling challenges in their hosting systems.
Submitters are asked to employ known workload description languages (e.g., SWF) to represent their traces or to attach schemas that allow interpreting them. The submission of artifacts that model the workloads is also encouraged.
Call for Open Scheduling Problems (OSP)
JSSPP welcomes novel, unpublished papers describing open problems in large-scale scheduling. Submissions must include a concise description of the key metrics of the system and how they are calculated, as well as anonymized data publication of the system workload and production schedule. Ideally, anonymized operational logs would also be published.
Submission Format
Paper formatting requirements are the same for Regular Papers, Workload Traces, and OSP-related submissions. Papers should be no longer than 20 single-spaced pages, 10pt font, including figures and references. All submissions must follow the LNCS format. See the instructions at Springer's website.
All papers will be reviewed by at least three members of the program committee.
Submissions are accepted via the EasyChair submission page.
Important Dates
- Paper Submission Deadline: February 24th, 2025 (FINAL)
- Author Notification: March 14th, 2025 (approximate, extended)
- Camera-Ready Paper: July 27th, 2025
تاریخهای کنفرانس
Conference Date
۱۳ خرداد ۱۴۰۴ → ۱۴ خرداد ۱۴۰۴
ارسال مقاله
Paper Submission Deadline
۶ اسفند ۱۴۰۳
اعلان
Author Notification
۲۴ اسفند ۱۴۰۳
نسخه نهایی
Camera-Ready Paper
۵ مرداد ۱۴۰۴
رتبه منبع
منبع: CORE2023
رتبه: C
حوزه پژوهشی: Distributed computing and systems software