Robust Real-Time Multiprocessor Interrupt Handling Motivated by GPUs

Elliott, Glenn A.; Anderson, James H.

doi:10.1109/ecrts.2012.20

Cited by 23 publications

(14 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Over the last few years, GPUs have started to gain attention in the real-time community [7,8,9,10,11]. Approaches to specific applications such as medical imaging have also been proposed [12].…”

Section: Related Work and Our Contri-butionsmentioning

confidence: 98%

Saving energy without defying deadlines on mobile GPU-based heterogeneous systems

Maghazeh

Bordoloi

Horga

et al. 2014

Proceedings of the 2014 International Conference on Hardware/Software Codesign and System Synthesis

View full text Add to dashboard Cite

With the advent of low-power programmable compute cores based on GPUs, GPU-equipped heterogeneous platforms are becoming common in a wide spectrum of industries including safety-critical domains like the automotive industry. While the suitability of GPUs for throughput oriented applications is well-accepted, their applicability for real-time applications remains an open issue. Moreover, in mobile/embedded systems, energy-efficient computing is a major concern and yet, there has been no systematic study on the energy savings that GPUs may potentially provide. In this paper, we propose an approach to utilize both the GPU and the CPU in a heterogeneous fashion to meet the deadlines of a real-time application while ensuring that we maximize the energy savings. We note that GPUs are inherently built to maximize the throughput and this poses a major challenge when deadlines must be satisfied. The problem becomes more acute when we consider the fact that GPUs are more energy efficient than CPUs and thus, a naive approach that is based on maximizing GPU utilization might easily lead to infeasible solutions from a deadline perspective.

show abstract

Section: Related Work and Our Contri-butionsmentioning

confidence: 98%

Saving energy without defying deadlines on mobile GPU-based heterogeneous systems

Maghazeh

Bordoloi

Horga

et al. 2014

Proceedings of the 2014 International Conference on Hardware/Software Codesign and System Synthesis

View full text Add to dashboard Cite

show abstract

“…First, we employ two token locks, one for long resources and the other for short resources. 7 Second, we replace Rule Q3 with the two rules below.…”

Section: Reducing Short-on-long Blockingmentioning

confidence: 99%

“…We do this by leveraging recent work on asymptotically optimal realtime k-exclusion protocols [7,8,13]. Such protocols provide a limited form of replication: they enable requests to be performed on k replicas of a single resource.…”

Section: Multi-unit Multi-resource Lockingmentioning

confidence: 99%

See 1 more Smart Citation

Fine-grained multiprocessor real-time locking with improved blocking

Ward

Anderson

2013

Proceedings of the 21st International Conference on Real-Time Networks and Systems

Self Cite

View full text Add to dashboard Cite

Existing multiprocessor real-time locking protocols that support nesting are subject to adverse blocking that can be avoided when additional resource-usage-pattern information is known. These sources of blocking stem from system overheads, varying critical section lengths, and a lack of support for replicated resources. In this paper, these issues are resolved in the context of the recently proposed realtime nested locking protocol (RNLP). The resulting protocols are the first to support fine-grained real-time lock nesting while allowing multiple resources to be locked in one atomic operation, both spin-and suspension-based waiting to be used together, and resources to be replicated. They also reduce "short-on-long" blocking, which is very detrimental if both very long and very short critical sections must be supported.

show abstract

“…The CUDA programming framework provides memory addressing technology that allows the GPU to directly access data allocated on the host for the purpose of mitigating the data transfer overhead, but this scheme is ill-suited for low-latency GPU computing due to the high cost of GPU's data access to the host memory. Given that GPUs are increasingly deployed in CPS applications [3,10,12,14], and real-time GPU resource management techniques are starting to be developed [1,2,5,6,7,8,9], it is time to look into a tighter integration of I/O processing and GPU computing.…”

Section: Introductionmentioning

confidence: 99%

Zero-copy I/O processing for low-latency GPU computing

Kato

Aumiller

Brandt

2013

Proceedings of the ACM/IEEE 4th International Conference on Cyber-Physical Systems

View full text Add to dashboard Cite

Cyber-physical systems (CPS) aim to monitor and control complex real-world phenomena where the computational cost and real-time constraints could be a major challenge. Manycore hardware accelerators such as graphics processing units (GPUs) promise to enhancing computation, leveraging the data parallelism often found in real-world scenarios of CPS, but performance is limited by the overhead of the data transfer between the host and the device memory. For example, plasma control in the HBT-EP Tokamak device at Columbia University [11,18] must execute the control algorithm in a few microseconds, but may take tens of microseconds to copy the data set between the host and the device memory. This paper presents a zero-copy I/O processing scheme that maps the I/O address space of the system to the virtual address space of the compute device, allowing sensors and actuators to transfer data to and from the compute device directly. Experiments using the plasma control system show a 33% reduction in computational cost, and microbenchmarks with more generic matrix operations show a 34% reduction, while in both cases, effective data throughput remains at least as good as the current best performers.

show abstract

Robust Real-Time Multiprocessor Interrupt Handling Motivated by GPUs

Cited by 23 publications

References 21 publications

Saving energy without defying deadlines on mobile GPU-based heterogeneous systems

Saving energy without defying deadlines on mobile GPU-based heterogeneous systems

Fine-grained multiprocessor real-time locking with improved blocking

Zero-copy I/O processing for low-latency GPU computing

Contact Info

Product

Resources

About