Dynamic task discovery in PaRSEC

Hoque, Reazul; Hérault, Thomas; Bosilca, George; Dongarra, Jack

doi:10.1145/3148226.3148233

Cited by 51 publications

(11 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…parameterized task graph approach requires users to express their computations in a subtle high-level formalism where the dataflow is explicitly described, but has less overhead on the runtime system. 27 Several efforts have been made to address the problem of reducing the overhead of task-based runtime systems, which mainly focus on those based on the sequential task flow model or increasing the amount of parallelism provided by these systems. In Reference 28, the authors analyzed the limiting factors in the scalability of a task-based runtime system and proposed individual solutions for each of the identified challenges, such as a wait-free dependency system and a scalable scheduler design based on delegation instead of work-stealing.…”

Section: Runtimementioning

confidence: 99%

Programming heterogeneous architectures using hierarchical tasks

Faverge

Furmento

Guermouche

et al. 2023

Concurrency and Computation

View full text Add to dashboard Cite

SummaryTask‐based systems have become popular due to their ability to utilize the computational power of complex heterogeneous systems. A typical programming model used is the Sequential Task Flow (STF) model, which unfortunately only supports static task graphs. This can result in submission overhead and a static task graph that is not well‐suited for execution on heterogeneous systems. A common approach is to find a balance between the granularity needed for accelerator devices and the granularity required by CPU cores to achieve optimal performance. To address these issues, we have extended the STF model in the StarPU runtime system by introducing the concept of hierarchical tasks. This allows for a more dynamic task graph and, when combined with an automatic data manager, it is possible to adjust granularity at runtime to best match the targeted computing resource. That data manager makes it possible to switch between various data layout without programmer input and allows us to enforce the correctness of the DAG as hierarchical tasks alter it during runtime. Additionally, submission overhead is reduced by using large‐grain hierarchical tasks, as the submission process can now be done in parallel. We have shown that the hierarchical task model is correct and have conducted an early evaluation on shared memory heterogeneous systems using the Chameleon dense linear algebra library.

show abstract

Section: Runtimementioning

confidence: 99%

Programming heterogeneous architectures using hierarchical tasks

Faverge

Furmento

Guermouche

et al. 2023

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…These runtimes build the DAG dynamically and unroll it as computational progress occurs. However, both runtimes suffer from sequential task insertion, which along with the DAG pruning phase, may have potential limitations on scalability [21]. There are numerous runtimes (HPX [22], Legion [23], Charm++ [24], etc.)…”

Section: Dynamic Runtime Systemsmentioning

confidence: 99%

Reshaping Geostatistical Modeling and Prediction for Extreme-Scale Environmental Applications

Cao

Abdulah

Alomairy

et al. 2022

SC22: International Conference for High Performance Computing, Networking, Storage and Analysis

Self Cite

View full text Add to dashboard Cite

“…It aims to shift the focus of developers from repetitive architectural details toward meaningful algorithmic improvements. Two domain specific languages are supported by Parsec, the Parameterized Task Graph [6] (PTG) and Dynamic Task Discovery [11] (DTD).…”

Section: Parsecmentioning

confidence: 99%

A Current Task-Based Programming Paradigms Analysis

Gurhem

Petiton

2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Task-based paradigm models can be an alternative to MPI. The user defines atomic tasks with a defined input and output with the dependencies between them. Then, the runtime can schedule the tasks and data migrations efficiently over all the available cores while reducing the waiting time between tasks. This paper focus on comparing several task-based programming models between themselves using the LU factorization as benchmark. HPX, PaRSEC, Legion and YML+XMP are task-based programming models which schedule data movement and computational tasks on distributed resources allocated to the application. YML+XMP supports parallel and distributed tasks with XscalableMP, a PGAS language. We compared their performances and scalability are compared to ScaLA-PACK, an highly optimized library which uses MPI to perform communications between the processes on up to 64 nodes. We performed a block-based LU factorization with the task-based programming model on up to a matrix of size 49512 × 49512. HPX is performing better than PaRSEC, Legion and YML+XMP but not better than ScaLAPACK. YML+XMP has a better scalability than HPX, Legion and PaRSEC. Regent has trouble scaling from 32 nodes to 64 nodes with our algorithm.

show abstract

Dynamic task discovery in PaRSEC

Cited by 51 publications

References 15 publications

Programming heterogeneous architectures using hierarchical tasks

Programming heterogeneous architectures using hierarchical tasks

Reshaping Geostatistical Modeling and Prediction for Extreme-Scale Environmental Applications

A Current Task-Based Programming Paradigms Analysis

Contact Info

Product

Resources

About