D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....only at the process level and a fixed process schedule. This causes a lack of adaptability and often necessitates an underutilization of available resources, problems that can be addressed with advanced real time communication services such as dynamic scheduling and adaptive resource management [21] [23] these are described later in this article) Furthermore, while these tools facilitate reuse at the component level, there are tremendous opportunities for also reusing common ways of integrating and reconfiguring components across similar control applications. These should be exploited. ....

....being received on the ground may need to be updated more often when a critical mission replanning task is being performed to avoid a threat. To support these needs, Boeing is currently extending the core OCP to incorporate recent advances in dynamic scheduling from Washington University [21] and adaptive resource management from Honeywell Laboratories [22] 23] By using dynamic scheduling algorithms in the event channel scheduler, the OCP can accommodate runtime changes in QoS requirements for events by allowing QoS specification changes to be specified and enforced at runtime. ....

[Article contains additional citation context not shown here]

D. Levine, C. Gill, and D. Schmidt, "Dynamic scheduling strategies for avionics mission computing," in Proc. 17th DASC. AIAA/IEEE/SAE Digital Avionics Systems Conf., 1998, vol. 1, pp. C15/1-8.

....correlation and pluggable scheduling strategies. NDDS, on the other hand, provides higher performance and online customizable replication mechanisms. Boeing is currently developing a Core OCP that uses RT CORBA and incorporates recent advances in dynamic scheduling from Washington University [16] and dynamic resource management from Honeywell Technologies [17, 18] An important feature of the Controls API we are generating at Georgia Tech is that it makes the underlying distribution mechanisms used by the Core OCP transparent to the controls system developer. This will enable a system ....

....the flexibility for rapid online adaptation and reconfiguration. There are a number of areas of ongoing work, including the following. Allowing runtime changes to quality of service properties of signals is a challenging open issue which will benefit from new dynamic scheduling algorithms [16] and dynamic resource management [17, 18] capabilities which Boeing is incorporating into the core OCP. A generic interface to mid level control components is being developed, which will provide support for hybrid control systems. This will provide common reconfiguration primitives used in fault ....

D. Levine, C. Gill, D. Schmidt. Dynamic Scheduling Strategies for Avionics Mission Computing. In Proc. 17th DASC. AIAA/IEEE/SAE. Digital Avionics Systems Conference, pp. C15/1-8, volume 1, 1998.

....providing system flexibility and modularity while maintaining real time performance. Dynamic scheduling. The current OCP prototype uses real time static scheduling in its event service. We plan to extend the OCP to take full advantage of recent advances in dynamic scheduling in the event service [6]. Unified reconfiguration process. We are exploring a unified reconfiguration process which closely couples software architectural reconfiguration with control reconfiguration. This will allow information needed by both types of reconfiguration to be shared (such as resource monitoring) This ....

D. Levine, C. Gill, D. Schmidt. Dynamic Scheduling Strategies for Avionics Mission Computing. In Proc. 17th DASC. AIAA/IEEE/SAE. Digital Avionics Systems Conference, pp. C15/1-8, volume 1, 1998.

No context found.

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....TAO, a high performance CORBA ORB, developed at Washington University St. Louis and University of California Irvine. TAO has been optimized for highperformance computing tasks[8] and has been successfully deployed in applications with stringent performance requirements such as flight avionics [9], telecommunications[11] and medical imaging[10] CORBA provides three methods for inter object communication: 1) a best effort one way invocation model, 2) a twoway synchronous request response model, and (3) a two way asynchronous method invocation (AMI) model. TAO comes with an ....

D. L. Levine, C. D. Gill, D. C. Schmidt. "Dynamic Scheduling Strategies for Avionics Mission Computing," Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998

....Real Time Middleware, Quality of Service Issues, Dynamic Scheduling Algorithms and Analysis, Adaptive Resource Management, Distributed Systems. 1 Introduction Next generation mission critical distributed real time and embedded (DRE) systems, such as integrated avionics mission computing systems [1], teams of collaborating emergency rescue robots [2] and distributed real time automobile management systems [3] must adapt swiftly to changing environmental conditions. Greater coordination allows elements at all levels to identify and respond effectively to transient opportunities and hazards. ....

....is also harmonic. In our current research, rate reallocations are controlled by a Real Time Adaptive Resource Manager (RTARM) 10] RT ARM is a middleware service developed by Honeywell that adapts the rates of tasks according to changing environmental conditions [11] In our previous research [1, 11], we specified that a task would have the same execution time across all rates. Our current research [12] however, has revealed uses for variable exe2 cution times across available rates. Most notably, we use variable execution times to provide finer granularity decomposition for execution of ....

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....tele immersion, immediate remote interaction with the physical environment can help maximize effectiveness at all levels of the system. For example, a group of UCAVs can share sensor data, post process data products, and remote operator requests. Next generation avionics mission computing systems [17], such as the sensor driven example shown in Figure 5, must collaborate with remote command and conI O Facade Sensor Proxy Sensor Proxy Sensor Proxy Sensor Proxy I O Facade I O Facade 2: Demarshaled data High Level Abstraction Low Level Abstraction 1: I O via interrupts Aircraft ....

....the real world, produce stringent requirements that serve to distill the key research challenges and design forces that must be addressed by QoS research to support these applications. The following design forces characterize the key research challenges we have identified based on our R D efforts [14, 1, 17, 19, 18, 20] developing next generation avionics mission computing systems. These forces must be addressed by researchers to ensure system correctness, performance, adaptability, and adequate resource utilization. Diverse inputs: Many next generation distributed applications must simultaneously use diverse ....

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....QoS requirements, these systems have become enabling technologies for companies competing in markets where deregulation and global competition motivate the need for increased software productivity, quality, and costeffectiveness. For instance, next generation avionics mission computing systems [2], such as the sensor driven example shown in Figure 1, must collaborate with remote command and control systems, provide on demand browsing capabilities for a human operator, and respond flexibly to unanticipated situational factors that arise in the run time environment [3] Moreover, these ....

....Center, Texas A M University, and the Georgia Institute of Technology. 2 2 Synopsis of Key Research Challenges and Design Forces The following design forces characterize the key research challenges we have identified based on our work developing mission critical adaptive real time systems [1, 8, 2, 6, 3, 21]. These forces must be addressed by researchers to ensure system correctness, adaptability, and adequate resource utilization. Diverse inputs: Mission critical systems must simultaneously use diverse sources of information, such as raw sensor data, command and control directives, and operator ....

[Article contains additional citation context not shown here]

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....in TAO borrow the client s thread of control to execute the servant s operation. Therefore, they are executed within the client thread at its thread priority. Although executing an operation in the client s thread is very efficient, it is undesirable for certain types of real time applications [28]. For instance, priority inversion can occur when a client in a lower priority thread invokes operations on a collocated object in a higher priority thread. To provide greater access control over the scope of TAO s collocation optimizations, therefore, applications can associate different access ....

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....factory that listens on a specific port number for clients to connect to the ORB instance running at a particular thread priority. TAO can be configured so that each priority level has its own Acceptor port. For instance, in statically scheduled, ratebased avionics mission computing systems [46], ports 10020, 10010, 10005, 10001 could be mapped to the 20 Hz, 10 Hz, 5 Hz, and 1 Hz rate groups, respectively. Requests arriving at these socket ports can then be processed by the appropriate fixed priority real time threads. Once a client connects, the Acceptor in the server ORB creates a new ....

....TAO s Run Time Scheduler consults a table of request priorities generated off line. At run time, TAO s ORB Core dispatches threads to the CPU(s) according to its dispatching mechanism. We are have extended TAO to support dynamically scheduling and applications with statistical QoS requirements [46]. 3.3 Efficient and Predictable Object Adapters The Object Adapter is the component in the CORBA architecture that associates a servant with an ORB, demultiplexes incoming client requests to the servant, and dispatches the appropriate operation of that servant. The key challenges associated with ....

[Article contains additional citation context not shown here]

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....ORBs have only recently begun specifying and implementing a standard [16] for composing and deploying componentizable services. Simultaneously, developers of high performance, real time, mission critical applications have begun employing distributed 6 object computing middleware based on CORBA [26, 33, 34, 35, 36, 37, 38, 24]. Therefore, we believe it is essential to start leveraging experience in designing, optimizing, and tuning QoS enabled ORB middleware to ensure standard CORBA Component Model (CCM) implementations will be sufficiently mature before they are considered for highperformance and real time systems. ....

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....QoS requirements, these systems have become enabling technologies for companies competing in markets where deregulation and global competition motivate the need for increased software productivity, quality, and costeffectiveness. For instance, next generation avionics mission computing systems [2], such as the sensor driven example shown in Figure 1, must collaborate with remote command and control systems, provide on demand browsing capabilities for a human operator, and respond flexibly to unanticipated situational factors that arise in the run time environment [3] Moreover, these ....

....real time mission critical systems; and Section 5 presents concluding remarks. 2 Synopsis of Key Research Challenges and Design Forces The following design forces characterize the key research challenges we have identified based on our work developing mission critical adaptive real time systems [1, 8, 2, 6, 3, 21]. 2 These forces must be addressed by researchers to ensure system correctness, adaptability, and adequate resource utilization. Diverse inputs: Mission critical systems must simultaneously use diverse sources of information, such as raw sensor data, command and control directives, and operator ....

[Article contains additional citation context not shown here]

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

.... to unanticipated situational factors that arise at run time (Doerr et al. 1999) Moreover, these systems must perform unobtrusively, shielding operators from unnecessary details, while simultaneously communicating and responding to mission critical information at an accelerated operational tempo (Levine et al. 1998). In such environments, it c fl 1999 Kluwer Academic Publishers. Printed in the Netherlands. 2 is often hard or impossible to predict a priori, or even approximate into the near future, system configurations or workload mixes. The communication infrastructure for these next generation systems ....

....programming power provided by QuO. 3.6. QOS SPECIFICATION IN QUO The QoS specification mechanisms provided by TAO, which were described in Section 3. 3, have been shown to be sufficient for building missioncritical applications with stringent real time requirements (Harrison et al. 1997; Levine et al. 1998; Gill et al. 2000; Doerr et al. 1999) However, these applications were built by researchers and developers with significant experi 23 ence and expertise both in real time systems and in distributed object oriented software engineering. It is a non trivial exercise for real time application ....

Levine, D. L., C. D. Gill, and D. C. Schmidt: 1998, `Dynamic Scheduling Strategies for Avionics Mission Computing'. In: Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC).

....constraints mentioned above, designers of missioncritical real time systems have also historically used relatively static method when allocating resources to system components that often possess competing real time requirements. For instance, flight qualified avionics mission computing systems [7, 8] establish the priorities for all resource allocation and scheduling decisions very early in the system lifecycle, i.e. well before run time. Static strategies have traditionally been used for mission critical real time applications because (1) system resources were insufficient for more ....

....was essential to remain on budget and on schedule, particularly when systems were designed from scratch using low level tools. The next generation of mission critical real time systems requires an increasingly wide range of features. For instance, next generation avionics mission computing systems [8, 7] must collaborate with remote command and control systems, provide on demand browsing capabilities for a human operator, and respond flexibly to unanticipated situational factors that arise in the run time environment. Moreover, these systems must perform unobtrusively, shielding human operators ....

[Article contains additional citation context not shown here]

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....in TAO borrow the client s thread of control to execute the servant s operation. Therefore, they are executed within the client thread at its thread priority. Although executing an operation in the client s thread is very efficient, it is undesirable for certain types of real time applications [24]. For instance, priority inversion can occur when a client in a lower priority thread invokes operations on a collocated object in a higher priority thread. To provide greater access control over the scope of TAO s collocation optimizations, therefore, applications can associate different access ....

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....of service (QoS) demands of nextgeneration real time applications requires object oriented (OO) middleware that is flexible, efficient, predictable, and convenient to program. Applications with deterministic realtime requirements, such as process control and avionics mission computing systems [1], impose severe constraints on the design and implementation of real time OO middleware. For example, avionics mission computing applications typically manage sensors and operator displays, navigate the aircraft s course, and control on board equipment. Middleware for such applications must ....

....from applications that use them. The architecture and behavior of TAO s strategized scheduling service is illustrated in Figure 7. This architecture evolved from our earlier work on a CORBA scheduling service [9] that supported purely static RMS for avionics mission computing applications [2, 1, 29]. Based on this work, as well as our experience prototyping dynamic scheduling strategies, we have identified the following set of common steps that are necessary to configure and process requests for a broad range of scheduling strategies: Step 1: A CORBA application specifies QoS information ....

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....invocations in TAO borrow the client s thread to execute the servant s operation. Therefore, they are executed within the client thread at its thread priority. Although executing an operation in the client s thread is very efficient, it is undesirable for certain types of real time applications [8]. For instance, priority inversion can occur when a client in a lower priority thread invokes operations on a collocated object that would otherwise be serviced by a higher priority thread. Therefore, to provide greater access control over the scope of TAO s collocation optimizations applications ....

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....regular stubs is completely transparent to clients. In TAO, collocated operation invocations borrow the client s thread to run the servant s operation. Therefore, they are executed with the client thread s priority. This design is undesirable, however, for certain types of real time applications [28] TAO TAO ORB C ORB COREORE OBJECT OBJECT ADAPTER ADAPTER CLIENT CLIENT OBJECT OBJECT ( SERVANTSERVANT) IDL IDL STUBS STUBS COLLOCATION TABLE COLLOCATION TABLE IIOP PROFILE IIOP PROFILE POA POA tango.cs:12345 tango.cs:54321 P P 2 2 P P 1 1 IDL IDL SKELETON SKELETON COLLOCATED COLLOCATED ....

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....factory that listens on a specific port number for clients to connect to the ORB instance running at a particular thread priority. TAO can be configured so that each priority level has its own Acceptor port. For instance, in statically scheduled, ratebased avionics mission computing systems [47], ports 10020, 10010, 10005, 10001 could be mapped to the 20 Hz, 10 Hz, 5 Hz, and 1 Hz rate groups, respectively. Requests arriving at these socket ports can then be processed by the appropriate fixed priority real time threads. Once a client connects, the Acceptor in the server ORB creates a new ....

....TAO s Run Time Scheduler consults a table of request priorities generated off line. At run time, TAO s ORB Core dispatches threads to the CPU(s) according to its dispatching mechanism. We are have extended TAO to support dynamically scheduling and applications with statistical QoS requirements [47]. 2.3 Efficient and Predictable Object Adapters The Object Adapter is the component in the CORBA architecture that associates a servant with an ORB, demultiplexes incoming client requests to the servant, and dispatches the appropriate operation of that servant. The key challenges associated with ....

[Article contains additional citation context not shown here]

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....Supporting the quality of service (QoS) demands of nextgeneration real time applications requires object oriented (OO) middleware that is flexible, efficient, predictable, and convenient to program. Applications with deterministic realtime requirements, such as avionics mission computing systems [1], impose severe constraints on the design and implementation of real time OO middleware. Avionics mission computing applications manage sensors and operator displays, navigate the aircraft s course, and control weapon release. Middleware for avionics mission computing must support applications ....

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....in TAO borrow the client s thread of control to execute the servant s operation. Therefore, they are executed within the client thread at its thread priority. Although executing an operation in the client s thread is very efficient, it is undesirable for certain types of real time applications [30]. For instance, priority inversion can occur when a client in a lower priority thread invokes operations on a collocated object in a higher priority thread. To provide greater access control over the scope of TAO s collocation optimizations, applications can associate different access policies to ....

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....factory that listens on a specific port number for clients to connect to the ORB instance running at a particular thread priority. TAO can be configured so that each priority level has its own Acceptor port. For instance, in statically scheduled, ratebased avionics mission computing systems [47], ports 10020, 10010, 10005, 10001 could be mapped to the 20 Hz, 10 Hz, 5 Hz, and 1 Hz rate groups, respectively. Requests arriving at these socket ports can then be processed by the appropriate fixed priority real time threads. Once a client connects, the Acceptor in the server ORB creates a new ....

....TAO s Run Time Scheduler consults a table of request priorities generated off line. At run time, TAO s ORB Core dispatches threads to the CPU(s) according to its dispatching mechanism. We are have extended TAO to support dynamically scheduling and applications with statistical QoS requirements [47]. 3.3 Efficient and Predictable Object Adapters The Object Adapter is the component in the CORBA architecture that associates a servant with an ORB, demultiplexes incoming client requests to the servant, and dispatches the appropriate operation of that servant. The key challenges associated with ....

[Article contains additional citation context not shown here]

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

....hashing [6] and active demultiplexing [7] optimizations to dispatch servant operations in constant O(1) time, regardless of the number of active connections, servants, and operations defined in IDL interfaces. 1 TAO s architecture for dynamically scheduled real time applications is described in [12]. R R U U N N T T I I M M E E S S C C H H E E D D U U L L E E R R SOCKET SOCKET QUEUEQUEUE DEMUXERDEMUXER ATM PORT INTERFACE CONTROLLER (APIC) Z Z E E R R O O C C O O P P Y Y B B U U F F F F E E R R S S I I OO SUBSYSTEM SUBSYSTEM OBJECT OBJECT ADAPTER ADAPTER SERVANT SERVANT ....

....factory that listens on a specific port number for clients to connect to the ORB instance running at a particular thread priority. TAO can be configured so that each priority level has its own Acceptor port. For instance, in statically scheduled, ratebased avionics mission computing systems [12], ports 10020, 10010, 10005, 10001 could be mapped to the 20 Hz, 10 Hz, 5 Hz, and 1 Hz rate groups, respectively. Requests arriving at these socket ports can then be processed by the appropriate fixed priority real time threads. Once a client connects, the Acceptor in the server ORB creates a new ....

[Article contains additional citation context not shown here]

D. L. Levine, C. D. Gill, and D. C. Schmidt, "Dynamic Scheduling Strategies for Avionics Mission Computing," in Proceedings of the 17th IEEE/AIAA Digital Avionics Systems Conference (DASC), Nov. 1998.

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