2005년 대한 생화학·분자생물학회 하계학술대회를 다녀와서

We consider the problem of fair resource allocation in a system containing different resource types, where each user may have different demands for each resource. To address this problem, we propose Dominant Resource Fairness (DRF), a generalization of max-min fairness to multiple resource types. We show that DRF, unlike other possible policies, satisfies several highly desirable properties. First, DRF incentivizes users to share resources, by ensuring that no user is better off if resources are equally partitioned among them. Second, DRF is strategy-proof, as a user cannot increase her allocation by lying about her requirements. Third, DRF is envy-free, as no user would want to trade her allocation with that of another user. Finally, DRF allocations are Pareto efficient, as it is not possible to improve the allocation of a user without decreasing the allocation of another user. We have implemented DRF in the Mesos cluster resource manager, and show that it leads to better throughput and fairness than the slot-based fair sharing schemes in current cluster schedulers.

https://static.usenix.org/event/nsdi11/tech/full_papers/Ghodsi.pdf

Dominant resource fairness: fair allocation of multiple resource types

Premises of creation of Internet portal designed to provide access to participants of educational and scientific process for the joint creation, consolidation, concentration and rapid spreading of educational and scientific information resources in its own depository are considered. CMS-based portal content management systems’ potentiality is investigated. Architecture for Internet portal of MES of Ukraine’s information resources is offered.

Створення порталу інформаційно-довідкових ресурсів міністерства освіти і науки україни

This survey covers hard real-time scheduling algorithms and schedulability analysis techniques for homogeneous multiprocessor systems. It reviews the key results in this field from its origins in the late 1960s to the latest research published in late 2009. The survey outlines fundamental results about multiprocessor real-time scheduling that hold independent of the scheduling algorithms employed. It provides a taxonomy of the different scheduling methods, and considers the various performance metrics that can be used for comparison purposes. A detailed review is provided covering partitioned, global, and hybrid scheduling algorithms, approaches to resource sharing, and the latest results from empirical investigations. The survey identifies open issues, key research challenges, and likely productive research directions.

/pdf/a-survey-of-hard-real-time-scheduling-for-multiprocessor-313o35g6ho.pdf

A survey of hard real-time scheduling for multiprocessor systems

In this chapter, you will see how the simplex method simplifies when it is applied to a class of optimization problems that are known as “network flow models.” You will also see that if a network flow model has “integer-valued data,” the simplex method finds an optimal solution that is integer-valued.

Flows in Networks

A new abstraction --- the Parallel Supply Function (PSF) --- is
 proposed for representing the computing capabilities offered by
 virtual platforms implemented atop identical multiprocessors. It is
 shown that this abstraction is strictly more powerful than
 previously-proposed ones, from the perspective of more accurately
 representing the inherent parallelism of the provided computing
 capabilities. Sufficient tests are derived for determining whether
 a given real-time task system, represented as a collection of
 sporadic tasks, is guaranteed to always meet all
 deadlines when scheduled upon a specified virtual platform using the
 global EDF scheduling algorithm.

/pdf/the-parallel-supply-function-abstraction-for-a-virtual-2i9s0cz178.pdf

The Parallel Supply Function Abstraction for a Virtual Multiprocessor

Mixed Criticality on MultiCore (MC2) is an ongoing project that explores the issue of providing operating system support for implementing mixed-criticality workloads upon multicore CPUs. Thus far, this project has primarily focused on implementation concerns - can mixed criticality be implemented upon multicores without incurring significant run-time implementation overhead? In this paper, we focus upon the workload model assumed in the MC2 framework: we propose several generalizations to the workload model and devise Integer Linear Programming (ILP) based scheduling and schedulability-analysis algorithms for the resulting generalized model.

https://dl.acm.org/doi/pdf/10.1145/3139258.3139271

An enhanced scheduler for MC2

The correctness of a real-time system depends on not only the system's output but also on the time at which results are produced. A hard real-time system is required to complete its operations before all its timing deadlines. For a given task set, it is very useful in an engineering context to know what changes to relative deadline can be made to a task that will deliver a schedulable system. In this paper, we address the sensitivity analysis (parameter calculations) of task relative deadline for EDF scheduled systems on a uniprocessor. We prove that a minimum task relative deadline can be determined by a single pass of the QPA algorithm. This algorithm provides exact and efficient sensitivity analysis for arbitrary deadline real-time systems. The approaches developed for task parameter calculations are therefore as efficient as QPA, and are easily incorporated into a system design support tool.

Notice of Retraction Sensitivity analysis of relative deadline for EDF scheduled real-time systems

We discuss the motivation behind, and the design and analysis of, an algorithm for synthesizing communication schedules for shared media networks in some safety-critical hard-real-time applications such as autonomous navigation and factory automation. Communication media may be inherently noisy in many such environments, and occasional transmission errors hence inevitable. Therefore it is essential that some degree of fault-tolerance be built into the communication protocol that is used --- in some safety-critical application domains, fault-tolerance requirements may be mandated by statutory certification requirements. Since the severity of the consequences of failing to successfully transmit different messages may be different, we consider a mixed-criticality setting in which the fault-tolerance requirement specification for messages are dependent on their criticality: more critical messages are required to be able to tolerate a larger number of faults. We advocate that communication schedules be "as static as possible" in safety-critical applications in order to facilitate verification and validation, and discuss the synthesis of semi-static schedules - schedules that are driven by precomputed lookup tables - with the desired fault-tolerance properties for such applications.

Fault-tolerant transmission of messages of differing criticalities across a shared communication medium

Temporal load-balancing, spreading out the executions of tasks over time, is desirable in many applications in complex systems A form of temporal load balancing is discussed, scheduling to maximize minimum inter-completion time (MICT-scheduling) and minimum global inter-completion time (MGICT-scheduling) It is shown that MICT- and MGICT-scheduling are, in general, NP-hard A number of restricted classes of task systems are identified, which can be efficiently MICT- and MGICT-scheduled

Sanjoy Baruah

Papers

The Parallel Supply Function Abstraction for a Virtual Multiprocessor

An enhanced scheduler for MC2

Notice of Retraction Sensitivity analysis of relative deadline for EDF scheduled real-time systems

Fault-tolerant transmission of messages of differing criticalities across a shared communication medium

Inter-completion time scheduling (ICTS): non-preemptive scheduling to maximize the minimum inter-completion time