JSSPP logo

The JSSPP workshop addresses all scheduling aspects of parallel processing, including cloud, grid (HPC) as well as “mixed/hybrid” or otherwise specific systems.

Large parallel systems have been in production for 25 years, creating the need of scheduling for such systems. Since 1995, JSSPP provides a forum for the research and engineering community working in the area. Initially, parallel systems were very static, with machines built in fixed configurations, which would be wholesale replaced every few years. Similarly, much of the workload was static as well, consisting of parallel scientific jobs running on a fixed number of nodes. Systems were primarily managed via batch queues. The user experience was far from interactive; jobs could wait in queues for days or even weeks.

A little over 10 years ago, the emergence of large scale, interactive, web applications together with the massive virtualization began to drive the development of a new class of (cloud) systems and schedulers. These systems would use virtual machines and/or containers to run "services", which would essentially never terminate (unlike scientific jobs). This created systems and schedulers with vastly different properties. Moreover, the enormous demand for computing resources resulted in a commercial market of competing providers. At the same time, the increasing demands for more power and interactivity have driven scientific platforms in a similar direction, causing the lines between these platforms to blur.

Nowadays, parallel processing is much more dynamic and connected. Many workloads are interactive and make use of variable resources over time. Complex parallel infrastructures can now be built on the fly, using resources from different sources, provided with different prices and quality of services. Capacity planning became more proactive, where resources are acquired continuously, with the goal of staying ahead of demand. The interaction model between job and resource manager is shifting to one of negotiation, where they agree on resources, price, and quality of service. Also, “hybrid” systems are often used, where the (virtualized) infrastructure is hosting a mix of competing workloads/applications, each having its own resource manager that must be somehow co-scheduled. These are just a few examples of the open issues facing our field.

From its very beginning, JSSPP has strived to balance practice and theory in its program. This combination provides a rich environment for technical debate about scheduling approaches including both academic researchers as well as participants from industry.

Building on this tradition, JSSPP welcomes both regular papers as well as descriptions of Open Scheduling Problems (OSP) in large scale scheduling. Lack of real-world data often substantially hampers the ability of the research community to engage with scheduling problems in a way that has real world impact. Our goal in the OSP venue is to build a bridge between the production and research worlds, in order to facilitate direct collaborations and impact.

Call for Regular Papers

JSSPP solicits papers that address any of the challenges in parallel scheduling, including:

For further information concerning paper formatting instructions please visit the Submission section.

Important dates related to Regular Papers:


Author Submission Deadline

February 10th, 2019 (extended)

Author Notification

March 10th, 2019

Workshop-Ready Paper

Workshop-ready paper: May 1st, 2019

Call for Open Scheduling Problems

NEW topic logo

Rationale and main goal:

One of the crucial roles of the JSSPP community is to enable the sharing, discussion, and possible resolution of scheduling problems. To fulfill this role, JSSPP also welcomes descriptions of Open Scheduling Problems (OSP) in datacenter and other large-scale environments.

Rules for submitting OSPs:

This call is primarily aimed at owners, operators, and clients of datacenters and large-scale environments. Each OSP, which must address a real or realistic situation, must include at least:

  1. the Scheduling Goals and their inter-relationship, including key metrics, SLOs, SLAs, etc., and to which stakeholder they are relevant and why,
  2. the structure of both the Workload and its Jobs, including key arrival and other operational patterns, job inter-dependency, elastic and moldable properties, etc.,
  3. the Environment to manage, including key size and topology information, types of resources, but also various operational and management rules such as scheduled maintenance, allocation and other constraints, etc.
Ideally, the OSP should also include:
  1. the actual workload logs,
  2. clear description of the current solution (solver) being used to solve the scheduling problem,
  3. scripts used in the real environment to compute various performance-related metrics used in that system
OSPs must be as clear as possible, to ensure the community addresses important problems robustly, but also to ensure reproducibility. To this end, the organizers provide a common human- and machine-readable format for presenting OSPs, and the basic tools associated with these [format and tools are available here]. The format will include extensibility options, to allow free-form descriptions, but authors of OSPs are encouraged to limit the use of these options to ensure reproducibility. The organizers are also well-aware of the importance of privacy and anonymity associated with the operation of datacenters and other large-scale environments, and will help authors of OSPs for this with advice based on own experience and with direct action if requested.


Prior to the workshop, the organizers will also engage the authors of accepted OSPs, and possibly ask for clarifications and more details about the proposed OSP, for example, the exact definition of a performance metric or SLO, the maintenance and failure patterns if not specified, other managerial goals that may have been reflected into the operation of the OSP environment. Ideally, OSPs will be accompanied by all the relevant open-access data and free open-source software, for example, anonymized operational logs and sanitized scripts used in the real environment to compute various metrics (iff. different from the tools provided by JSSPP). Description of the current solution (solver) being used to solve the presented scheduling problem is also welcome.

Publication options:

We accept two different publication options for OSPs:

  1. regular papers, where the authors must attend the JSSPP workshop and the OSPs are later included in the JSSPP proceedings and also published on the website.
  2. OSPs are discussed during the JSSPP 2019 workshop. The authors are not required to attend JSSPP, yet their presence is most welcome. The OSPs are published on the JSSPP website, and later in an arXiv technical report.
In both cases, data and software will be published on relevant repositories, such as Zenodo and GitHub, respectively. For further information concerning paper formatting instructions please visit the Submission section.

Important dates related to OSP:


Format and basic tools online

December 18th, 2018: [format and tools]

OSP Submission Deadline

February 10th, 2019 (extended)

Author Notification

February 15th, 2019

Public List of Accepted OSPs

March 1st, 2019

OSPs presented at JSSPP

May 24th, 2019

Tech. Report/Papers Published

Fall 2019

JSSPP 2019 - the Workshop on Jobs Scheduling Strategies for Parallel Processing. Contact email: jssppw@gmail.com