The Open Community Runtime on the Intel Knights Landing Architecture

Dokulil, Jiri; Benkner, Siegfried; Yaghob, Jakub

doi:10.1007/978-3-319-65482-9_65

Cited by 4 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There is a significant performance penalty when a CPU core access memory that belongs to a different NUMA node. The actual performance impact depends heavily on the application, ranging from negligible to over 2x slowdown (as observed in both our and independent experiments [19,34]). Most operating systems employ heuristics to place (and possibly also move) application data in memory in a way that improves locality.…”

Section: Numa Supportsupporting

confidence: 68%

“…We have started looking at NUMA in OCR when experimenting on the Intel Knights Landing architecture [19], which provides on-chip high-bandwidth memory and optional NUMA modes. We were only running the applications on one machine, using shared memory.…”

Section: Numa Supportmentioning

confidence: 99%

“…The architectures are as follows: a dual socket server with 8 core CPUs (2x8), the Xeon Phi coprocessor (KNC), the Xeon Phi processor with NUMA disabled (KNL), the Xeon Phi processor with 4 NUMA nodes (KNL SNC4), and four socket server with 20 core CPUs (4x20). The data are a compilation of older experiments [19,24] and new results. It shows that performance of OCR is competitive when compared with OpenMP and TBB.…”

Section: Shared-memory Systemsmentioning

confidence: 99%

“…Since 2014, we have been developing an implementation of the OCR specification called OCR-Vx [18,24]. We have targeted both shared-memory and distributed-memory parallel architectures, as well as some less common architectures like the Intel Xeon Phi (both in the KNC coprocessor [18] and KNL processor format [19]) and OpenCL accelerators. We have also made multiple contributions to the OCR specification and developed a detailed consistency model for OCR [15].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The OCR-Vx experience: lessons learned from designing and implementing a task-based runtime system

Dokulil

Benkner

2022

J Supercomput

Self Cite

View full text Add to dashboard Cite

Task-based runtime systems are an important branch of parallel programming research, since tasks decouple computation from the compute units, giving the runtime systems greater flexibility than a thread-based solution. This makes it easier to deal with the ever-increasing complexity of parallel architectures by providing a separation of concerns—the specification of parallelism is separated from the implementation of the parallel computations on a specific architecture. The Open Community Runtime is one such system, aimed at large-scale parallel systems. Unlike many other task-based runtime systems, the creators not only provided an implementation but there is also a comprehensive specification document. This has allowed us to create an independent implementation, called OCR-Vx. In this article, we present our experience of developing the runtime system, put our work in the context of the specification and the other implementations, and describe key lessons that we have learned during our work. We discuss the design and implementation issues of task-based runtime systems and applications including task synchronization and scheduling, data management, memory consistency, the relation between shared-memory and distributed-memory runtime systems, NUMA architectures, and heterogeneous systems. The article is aimed at audiences not familiar with OCR, since we believe these lessons could be valuable for developers working on other task-based runtime systems or designing new ones.

show abstract

Section: Numa Supportsupporting

confidence: 68%

Section: Numa Supportmentioning

confidence: 99%

Section: Shared-memory Systemsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

The OCR-Vx experience: lessons learned from designing and implementing a task-based runtime system

Dokulil

Benkner

2022

J Supercomput

Self Cite

View full text Add to dashboard Cite

show abstract

“…As a result, the existing implementation does not have to be changed to implement the proposed object model. The reference implementation created by Intel/Rice University [16], a derived implementation from PNNL [14], and our OCR-Vx implementation [5][6][7] all already conform to the proposed object model. In some cases, they may provide stronger guarantees than required, so the model gives them room for optimizations.…”

Section: Introductionmentioning

confidence: 99%

Consistency model for runtime objects in the Open Community Runtime

Dokulil

2018

J Supercomput

Self Cite

View full text Add to dashboard Cite

Task-based runtime systems are seen as a way to deal with increasing heterogeneity and performance variability in modern high performance computing systems. The Open Community Runtime is an example of such runtime system, with the benefit of having an open specification that defines its interface and behavior. This allows others to study the design and create alternative implementations of the runtime system. However, it turns out that the consistency model that defines how the OCR programs are synchronized when executed in parallel is not sufficiently defined. In the following text, we complete the consistency model by filling in the missing pieces and extending the memory model given in the specification.

show abstract

Adaptive Scheduling of Collocated Applications Using a Task-Based Runtime System

Dokulil

Benkner

2018

2018 30th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD)

View full text Add to dashboard Cite

The Open Community Runtime on the Intel Knights Landing Architecture

Cited by 4 publications

References 7 publications

The OCR-Vx experience: lessons learned from designing and implementing a task-based runtime system

The OCR-Vx experience: lessons learned from designing and implementing a task-based runtime system

Consistency model for runtime objects in the Open Community Runtime

Adaptive Scheduling of Collocated Applications Using a Task-Based Runtime System

Contact Info

Product

Resources

About