On-line dependability enhancement of multiprocessor SoCs by resource management

Braak, Timon D. ter; Burgess, Stephen T.; Hurskainen, Heikki; Kerkhoff, Hans G.; Vermeulen, Bart; Zhang, X.

doi:10.1109/issoc.2010.5625564

Cited by 15 publications

(11 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• Fault detection mechanisms A rich literature with methods to test the modules of the processing elements is available, with approaches adopted at different levels or modules. Fault detection at the system level [4,19], fault detection at the router level [13,30,31], fault detection at the processor level [5,28]. Our proposal can adopt these mechanisms since fault detection is out of the scope of this work.…”

Section: (A) (B)mentioning

confidence: 99%

See 1 more Smart Citation

System management recovery in NoC-based many-core systems

Silva

Caimi

Moraes

et al. 2020

Analog Integr Circ Sig Process

View full text Add to dashboard Cite

The technology nodes reduction enabled the emergence of NoC-based many-cores with dozens to hundreds of processing elements (PEs). Despite the processing power offered by a large number of processors and communication flexibility due to the adoption of NoCs, it is necessary to manage the many-core resources to ensure scalability. The execution of the management tasks requires a PE reserved exclusively to execute such actions. These processors are named managers PE-MPE. A centralized approach would induce a significant load to the MPE in large-scale systems, and a permanent fault in the MPE would compromise the entire system. The adoption of a distributed approach, organization adopted in this work, with MPEs hierarchically organized, reduces the management load, and a fault in an MPE would compromise only the PEs managed by the faulty MPE. The literature presents several fault-tolerant proposals targeting the NoC or the processors. However, there is a significant gap related to fault-tolerant methods at the system level, i.e., related to fault-tolerant techniques regarding the MPEs. The goal of this paper is to present a recovery method when an MPE became faulty, and propose a protocol to migrate the management software safely to a new PE. The method adopts task migration to release a processor if there is no processor to receive the kernel that was executing in a faulty processor. The proposal is transparent to the applications running in the many-core, with an overhead in the execution time varying between 1.5 and 1.65 ms during the management and task migration.

show abstract

Section: (A) (B)mentioning

confidence: 99%

“…In this example, it is necessary to migrate the task executing on SP7 to SP1. The CM0 sends a message to SP7 to migrate the task it is executing to SP1 (5). As detailed in Sect.…”

Section: Recovery Protocol Stepsmentioning

confidence: 99%

System management recovery in NoC-based many-core systems

Silva

Caimi

Moraes

et al. 2020

Analog Integr Circ Sig Process

View full text Add to dashboard Cite

show abstract

“…The platform model is then updated according to the latest test results. More detailed information about this dependability approach is described in [22].…”

Section: B Platform Discovery and Fault Detectionmentioning

confidence: 99%

Adaptive resource allocation for streaming applications

Braak

Toersche

Kokkeler

et al. 2011

2011 International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation

View full text Add to dashboard Cite

Streaming applications often have latency and throughput requirements due to timing critical signal processing, or the time critical interaction with their environment. Mapping such applications to a multi-core architecture is commonly done at design-time to be able to analyze the complex design-space. However, such design-flows cannot deal with a dynamic platform or a dynamic set of applications. Hardware faults and resources claimed by other applications may render the assumed available resources inaccessible. To avoid the assumptions posed on the state of the platform by a fixed resource allocation, applications should be designed with location transparency in mind. Applications must be analyzed at design-time to determine the required resource budget, independent of which specific resources will be allocated. Sufficient performance can be guaranteed when such applications are mapped onto an architecture in which each resource is arbitrated using a budget scheduler.Within the Cutting edge Reconfigurable ICs for Stream Processing (CRISP) project, a many-core platform is developed that adheres to these requirements. Using the configuration features of the platform, the system is able to control at run-time what resources are being used by the applications. This paper shows that run-time resource allocation can effectively adapt to the available set of resources, providing partial distribution transparency to the user. As an example, a GNSS receiver is mapped to the platform containing faulty hardware components. A few resources remain critical, but in most cases the faulty components can be circumvented, such that adequate resources can be allocated to the application at run-time.

show abstract

“…The dependability issue in the context with NoCs is raising interest in research in the recent years. In [7] an approach for online resource management for MPSoCs is presented and [8] presents a spare router approach. Universal dependability concepts are still less explored.…”

Section: Dependable Soc Communication Architecturesmentioning

confidence: 99%

Invited paper: Design criteria for dependable System-on-Chip architectures

Hollstein

Samman

Jaiswal

et al. 2011

6th International Workshop on Reconfigurable Communication-Centric Systems-on-Chip (ReCoSoC)

View full text Add to dashboard Cite

The rapid improvement of semiconductor technologies is the enabling factor for the design of large-scale Systemon-Chip (SoC) architectures. At the same time the scale-down of feature sizes in silicon technologies brings up new challenges as parameter variations of the transistor devices, an increased vulnerability for wear-out effects during the lifetime of the device and increased sensitivity for soft-errors. The overall system reliability is therefore an important topic to be addressed in the SoC design process as well as economic considerations related to manufacturing yield and lifetime maintainability. The aim of this contribution is to outline implications for the design process and to illustrate the dependability aspects at the example of SoC communication architectures being implemented as Networkson-Chip (NoCs).

show abstract

On-line dependability enhancement of multiprocessor SoCs by resource management

Cited by 15 publications

References 24 publications

System management recovery in NoC-based many-core systems

System management recovery in NoC-based many-core systems

Adaptive resource allocation for streaming applications

Invited paper: Design criteria for dependable System-on-Chip architectures

Contact Info

Product

Resources

About