Nonblocking Epochs in MPI One-Sided Communication

Zounmevo, Judicael A.; Zhao, Xin; Balaji, Pavan; Gropp, William; Afsahi, Ahmad

doi:10.1109/sc.2014.44

Cited by 6 publications

(5 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, this lack of nonblocking variants collides with the asynchronous philosophy of our proposal. This topic is not new; Zounmevo et al [15] propose an MPI extension with non-blocking synchronizations. By standardizing these non-blocking variants, TAMPI could directly support all RMA modes.…”

Section: Task-aware Rma Communicationsmentioning

confidence: 99%

“…By standardizing the non-blocking synchronizations [15] and remote notifications [8], TAMPI could provide a finegrained RMA integration with tasking models without any significant change on the library. However, in this paper, we design and implement our approach based on the GASPI model [10] because it already provides all these properties.…”

Section: Task-aware Rma Communicationsmentioning

confidence: 99%

See 1 more Smart Citation

Combining One-Sided Communications with Task-Based Programming Models

Sala

Macià

Beltrán

2021

2021 IEEE International Conference on Cluster Computing (CLUSTER)

View full text Add to dashboard Cite

Hybrid programming combining task-based and message-passing models is an increasingly popular technique to exploit multi-core clusters. The Task-Aware MPI (TAMPI) library integrates both models enabling the safe overlap of computation and communication tasks using two-sided MPI communications. Two-sided primitives combine data transfers with implicit synchronizations, but one-sided models usually offer more efficient data transfers decoupling synchronizations. MPI offers four distinct one-sided synchronization modes, while GASPI is a PGAS API providing one-sided operations with remote notifications for fine inter-process synchronizations.In this paper, we study the challenges of integrating MPI and GASPI one-sided operations with the OpenMP and OmpSs-2 tasking models. We propose and implement several extensions to the GASPI and OmpSs-2 programming models, which are leveraged by a new library called Task-Aware GASPI (TAGASPI). The TAGASPI library allows the efficient and safe use of one-sided operations with remote notifications inside tasks. Both TAGASPI and TAMPI transparently manage communications issued by tasks and allow these to overlap with computation tasks naturally, following a data-flow model. These libraries are complementary and can be mixed in the same application.Our experience porting several mini-apps to this hybrid model shows that TAGASPI helps leverage one-sided communications with similar complexity to pure and hybrid two-sided MPI approaches. We show that our hybrid one-sided approach outperforms the pure MPI strategies, but it also surpasses the TAMPI's performance when stressing communication phases, e.g., increasing the communication parallelism and reducing the communication tasks' sizes.

show abstract

Section: Task-aware Rma Communicationsmentioning

confidence: 99%

Section: Task-aware Rma Communicationsmentioning

confidence: 99%

Combining One-Sided Communications with Task-Based Programming Models

Sala

Macià

Beltrán

2021

2021 IEEE International Conference on Cluster Computing (CLUSTER)

View full text Add to dashboard Cite

show abstract

“…With the design of SCOSCo, it is shown how this approach is also extensible to non‐cache‐coherent systems. Nevertheless, there has been criticism to the MPI application program interface design, which has been addressed by the standard and in publications …”

Section: Related Workmentioning

confidence: 99%

“…Nevertheless, there has been criticism to the MPI application program interface design, which has been addressed by the standard and in publications. 4,[18][19][20][21]…”

Section: Related Workmentioning

confidence: 99%

Exploring one‐sided communication and synchronization on a non‐cache‐coherent many‐core architecture

Christgau

Schnor

2017

Concurrency and Computation

View full text Add to dashboard Cite

Summary The ongoing many‐core design aims at core counts where cache coherence becomes a serious challenge. Therefore, this paper discusses how one‐sided communication and the required process synchronization can be realized on a non‐cache‐coherent many‐core CPU. The Intel Single‐chip Cloud Computer serves as an exemplary hardware architecture. The presented approach is based on software‐managed cache coherence for MPI one‐sided communication. The prototype implementation delivers a PUT performance of up to 5 times faster than the default message‐based approach and reveals a reduction of the communication costs for the NAS Parallel Benchmarks 3‐D fast Fourier Transform by a factor of 5. Further, the paper derives conclusions for future non‐cache‐coherent architectures.

show abstract

“…With the design of SCOSCo, we have shown how this approach is also extensible to non-cache coherent systems. Nevertheless, there has been criticism to the MPI API design which has been addressed by the standard and in publications [2,3,24,26,31].…”

Section: Related Workmentioning

confidence: 99%

Software-managed Cache Coherence for fast One-Sided Communication

Christgau

Schnor

2016

Proceedings of the 7th International Workshop on Programming Models and Applications for Multicores and Manycores

View full text Add to dashboard Cite

The ongoing many-core design aims at core counts where cache coherence becomes a serious challenge. Therefore, this paper discusses how one-sided communication can be implemented on a non-cache coherent many-core CPU. The Intel SCC serves as an exemplary hardware architecture. The presented approach is based on software-managed cache coherence for MPI one-sided communication. The prototype implementation delivers a PUT performance of up to five times faster than the default message-based approach and reveals a reduction of the communication costs for the NPB 3D FFT by a factor of five. Further, the paper identifies drawbacks of the SCC's architecture and derives conclusions for future architectures.

show abstract

Nonblocking Epochs in MPI One-Sided Communication

Cited by 6 publications

References 14 publications

Combining One-Sided Communications with Task-Based Programming Models

Combining One-Sided Communications with Task-Based Programming Models

Exploring one‐sided communication and synchronization on a non‐cache‐coherent many‐core architecture

Software-managed Cache Coherence for fast One-Sided Communication

Contact Info

Product

Resources

About