J. Stankovic and K. Ramamrithm. The Spring Kernel: A New Paradigm for Real-Time Systems. IEEE Software, 8(3), May 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....needs to access multiple resources, additional information regarding dependencies among tasks, each corresponding to a resource usage, is needed. IRS [5] requires the application programmers to specify a task precedence graph through a special system call interface. In contrast, the Spring system [15] is able to derive this information automatically through a compiler, but requires that all applications be written using a special Spring C programming language. As realtime operating systems incorporate more sophisticated multiresource coordination mechanisms, more application level information ....

J. Stankovic and K. Ramamritham. The Spring Kernel: A new paradigm for real-time systems. IEEE Software, 8(3), May 1991. 5

....real time requirement, meaning to guarantee to schedule all tasks on arrival such that they are executed in time with full functionality. The question is how to soften the requirement such that the remaining guarantee is acceptable. More recently, so called dynamic scheduling has been proposed [3, 4]. It allows the scheduler to decline the scheduling of an arrived task, if it cannot guarantee meeting its deadline. Based on a given worst case execution time (WCET) dynamic scheduling provides predictability for all tasks accepted for scheduling and subsequent execution. Still, the problem ....

J. Stankovic and K. Ramamdtham: "The Spring Kernel: A New Paradigm for Real-Time Systems", 1EEE Software, 8(3), pp. 62-72, May 1991

.... F18 Multiple suitable resources Distributed spanning endsystems F19 Heterogeneity among operating Distributed systems and endsystems local scheduling F20 Dynamic heterogeneous Distributed applications within an endsystem F21 Competing quality of service Distributed requirements F22 Abstract state consistency Distributed F23 Temporal consistency Distributed The key insight offered by this example is that the priority inversion problem emerges from the combination of: 1) the fundamental design force that some rational allocation of resources is necessary to ensure ....

....timeliness may require sacrificing accuracy, or vise versa. Similarly, tradeoffs arise among most QoS requirement categories:timeliness and security, security and accuracy, timeliness and reliability. Distributed scheduling decisions should consider global QoS requirements and tradeoffs. F22: Abstract state consistency. While it is not possible to maintain a completely accurate, up to date picture of the global state of an entire distributed system, it is necessary to maintain a view that is consistent, i.e.fundamental properties such as causality are not violated, within some level ....

[Article contains additional citation context not shown here]

J. A. Stankovic and K. Ramamritham, "The Spring Kernel: A New Paradigm for Real-Time Systems," IEEE Software, vol. 8, pp. 62-- 72, May 1991.

....scheduling and analysis theory. This causes prob lems when new ideas need to be tested and evalu ated, since the real time aspect differs between stateof the practice and state of the art. There already ex ists state of the art RTOS, for example MiThos [9] the Spring kernel [15] and Emeralds [19] all with dif ferent advantages and drawbacks. These systems are often not general solutions but aimed at something special, for example special hardware or scheduling theory. Most of the state of the practice (commercial, off the shelf) RTOS are in some way configurable in or ....

....can easily connect interrupts with tasks thus creating an aperiodic task (waiting for a signal interrupt) Although a handler is supported by the kernel, a user can rewrite it to support their specific system (interrupt controlled devicedrivers etc. This solution has already been implemented [15] [2] and the result is satisfactory [1] The kernel does not fully support interrupts, since this is hardware specific and not yet supported by the Obelix Configuration Tool, see section 8. A pseudocode of the default interrupt handler is shown in figure 5.4. 5.4 Interrupts The interrupt ....

J. A. Stankovic and K. Ramamritham. The spring kernel: A new paradigm for real-time systems. IEEE, 1991.

....developing support for fault tolerance, investigating object oriented approaches, providing support for multiprocessor and distributed real time systems including end to end timing constraints, and attempting to define a real time micro kernel. Several projects, such as MARS[5] Spring[6], ARTS[7] HARTOS[8] and so on, are of this type of research. But, they focus on specific issues, and the issues which are not their focus are often weak on support. In this paper, we present the design and implementation of a real time environment at user level. The real time thread package ....

J. A. Stankovic, and K. Ramamritham. The Spring kernel: A new paradigm for real-time systems. IEEE Software, vol. 8, no. 3, pp. 62-72, May 1991.

....of each aperiodic task to determine whether or not it can meet its deadline. Examples include, aperiodic response time minimization [11] slack maximization [4] slack stealing [23] the reservation based (RB) algorithm [2] and the guarantee routines introduced most notably in the Spring kernel [21]. The utilization bound described in this paper is the first constant time test that enables us to efficiently determine the schedulability of aperiodic workloads on multiprocessors. 7 Conclusions In this paper, we derived, for the first time, the optimal utilization bound for the schedulability ....

J. A. Stankovic and K. Ramamritham. The Spring Kernel: A new paradigm for real-time systems. IEEE Software, pages 62--72, May 1991.

.... control has been used to guarantee predictability of services where request patterns are not known in advance, e.g. establishment re quests of real time channels [5] This concept has also been applied to resource reservation for dynamically arriving real time tasks, e.g. the Spring Kernel [6] and Dreams real time system [7] A main concern of this approach is predictability. Run time guarantees given to admitted requests are never revoked even if they resuit in rejecting subsequently arriving more important requests competing for the same resources. In soft real time systems, ....

J. A. Stankovic and K. Ramamritham, "The Spring Kernel: A new paradigm for real-time systems, " IEEE Software, pp. 62 72, May 1991.

.... providing guarantees in performance critical systems has been to rely on static allocation and scheduling algorithms that assume full a priori knowledge of the resource requirements of tasks [5, 31, 32, 40, 41] The concept of dynamic real time systems [34] pioneered by the Spring kernel project [35, 36], was introduced later to describe applications where runtime workload is unknown until admission control time. It resulted in innovative planning based scheduling algorithms that provide online guarantees for dynamically arriving tasks [18, 25, 28, 33, 37, 43, 44] Task execution times where ....

J. A. Stankovic and K. Ramamritham. The Spring Kernel: A new paradigm for real-time systems. IEEE Software, pages 62--72, May 1991.

....schedulability analysis, several approaches have been proposed in the literature. In [8] resource contention is solved o line by constructing a static schedule, which is stored in a table and enforced at runtime using a time driven approach. A di erent approach is used in the Spring kernel [15], where the schedule is constructed using a heuristic function which may integrate timing, resource, and precedence constraints [21] Sha, Rajkumar, and Lehoczky, in [14] proposed two concurrency control protocols, the Priority Inheritance Protocol (PIP) and the Priority Ceiling Protocol (PCP) ....

J.A. Stankovic and K. Ramamritham, The Spring Kernel: A New Paradigm for Real-Time Systems, IEEE Software, Vol. 8, No. 3, pp. 62-72, (May 1991).

....real time tasks are independent. On the other hand, some of the solutions proposed in the literature for handling resource constraints did not consider hybrid (hard and soft) task sets, but developed the analysis only for homogeneous sets consisting of hard periodic tasks [15, 2] or sporadic tasks [20, 16, 8]. A method for analyzing the schedulability of hybrid task sets where hard tasks may share resources with soft tasks handled by dynamic aperiodic servers was presented by Ghazalie and Baker in [7] Their approach is based on reserving an extra budget to the aperiodic server for synchronization ....

J.A. Stankovic and K. Ramamritham, The Spring Kernel: A New Paradigm for Real-Time Systems, IEEE Software, Vol. 8, No. 3, pp. 62-72, (May 1991).

....we have demonstrated in this paper, a xed slack sharing scheme can lead to wide load imbalance between di erent resources, leading to fewer number of applications admitted. Further, the e ect of precedence constraints with other tasks in the real time application is not addressed. Spring Kernel [26] provides real time support for multiprocessor and distributed environments using dynamic planning based scheduling. A computation is automatically broken into precedence related tasks and the worst case execution time is automatically derived from source code analysis. While Spring aims for ....

J. Stankovic and K. Ramamritham, \The Spring Kernel: A New Paradigm for Real-time Systems", IEEE Software, Vol. 8, No. 3, 62-72, May 1991.

....the threads package enables both the spawning and spawned thread to be executed under hard real time constraints. The most important property of UMass Spring threads is its run time predictability. This predictability is ensured through the support and scheduling model of the UMass Spring kernel [2]. The UMass Spring kernel uses a dynamic, planning based approach to resource usage, thus avoiding the blocking on resources that occurs in systems that are priority based. This paper describes the design of the UMass Spring threads package and its implementation and measurement on a ....

....occurs in systems that are priority based. This paper describes the design of the UMass Spring threads package and its implementation and measurement on a representative platform. The real time threads package builds upon previous work in the design and implementation of the UMass Spring kernel [2]. This paper extends the presentation of the high level design and initial implementation contained in [3] This completed implementation is, to our knowledge, the first threads package with such rich semantics that is suitable for hard real time environments. The UMass Spring threads package ....

[Article contains additional citation context not shown here]

John A. Stankovic and Krithi Ramamritham, "The Spring kernel: A new paradigm for real-time systems," IEEE Software, vol. 8, no. 3, pp. 62--72, May 1991.

No context found.

J. Stankovic and K. Ramamrithm. The Spring Kernel: A New Paradigm for Real-Time Systems. IEEE Software, 8(3), May 1991.

No context found.

J.A.Stankovic, K.Ramamritham, "The Spring Kernel: A New Paradigm for Real-Time Systems ", IEEE Software, Vol. 8, No. 3, 1991, pp. 62-72.

No context found.

J.A.Stankovic, K.Ramamritham, \The Spring Kernel: a New Paradigm for Real Time systems, " IEEE Software, Vol.8, No.3, pp.62-72, may 1991.

No context found.

Stankovic, J. A., and Ramamritham, K. The spring kernel: a new paradigm for real-time systems. IEEE Software 8, 3 (May 1991), 62-72.

No context found.

J.A. Stankovic and K. Ramamritham, "The Spring Kernel: A New Paradigm for Real-Time Systems," IEEE Software, vol. 8, no. 3, pp. 62-72, May 1991.

No context found.

J. Stankovic and K. Ramamritham, "The Spring Kernel: A New Paradigm for Real-Time Systems," IEEE Software, vol. 8, no. 3, pp. 62-72, May 1991.

No context found.

J. A. Stankovic and K. Ramamritham. The Spring Kernel: A new paradigm for real-time systems. IEEE Software, pages 62--72, May 1991.

No context found.

J.A. Stankovic and K. Ramamritham, "The Spring Kernel: A New Paradigm for Real-Time Systems," IEEE Software, pp. 62-72, May 1991.

No context found.

J. Stankovic and K. Ramamrithm. The Spring Kernel: A New Paradigm for Real-Time Systems. IEEE Software, 8(3), May 1991.

No context found.

J. A. Stankovic and K. Ramaritham,"The Spring Kernel: a new paradigm for real-time systems", IEEE Software, vol.8, no.3, pp.62-72, May 1991.

No context found.

J. A. Stankovic and K. Ramamritham. The Spring Kernel: A new paradigm for real-time systems. IEEE Software, pages 62--72, May 1991.

No context found.

J.A. Stankovic and K.Ramamritham, "The Spring Kernel: A New Paradigm for Real-time Systems," IEEE Software, Vol. 8, No. 3, pp. 6272, May 1991.

No context found.

J. Stankovic and K. Ramamritham, "The Spring Kernel: A New Paradigm for Real-time Systems," IEEE Software, Volume 8, Number 3, pages 6272, May 1991.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC