Providing high‐level self‐adaptive abstractions for stream parallelism on multicores

Vogel, Adriano; Griebler, Dalvan; Fernandes, Luiz Gustavo

doi:10.1002/spe.2948

Cited by 11 publications

(9 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In parallel computing, many entities can be adapted at run-time [13], such as batches size [19] for increasing throughput, the cores frequency for reducing energy consumption [5] and dynamically changing the degree of parallelism [4,15,23]. However, these optimizations are arguably not flexible enough for the adaptations that real-world stream processing applications demand.…”

Section: Results Summarymentioning

confidence: 99%

“…If not, goes to step 3. The decision-making strategy infers that two values are significantly different when they contrast higher than 20% (a threshold ), which is a configurable parameter that the used value of 20% was ascertained in previous work [23]. 3.…”

Section: Decision-making Strategymentioning

confidence: 99%

“…Moreover, the continuous data arrival requires stream processing applications to run for long or infinite periods, where such long executions are often subject to fluctuations in the environment (e.g., resource availability) and at the application level (input rate, workload) [24]. Hence, self-adapting entities (e.g., degree of parallelism, cores, and their frequencies) during the execution is important for achieving responsiveness [4,5,15,23].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Online and transparent self-adaptation of stream parallel patterns

et al. 2021

Self Cite

View full text Add to dashboard Cite

Several real-world parallel applications are becoming more dynamic and long-running, demanding online (at run-time) adaptations. Stream processing is a representative scenario that computes data items arriving in real-time and where parallel executions are necessary. However, it is challenging for humans to monitor and manually self-optimize complex and long-running parallel executions continuously. Moreover, although high-level and structured parallel programming aims to facilitate parallelism, several issues still need to be addressed for improving the existing abstractions. In this paper, we extend self-adaptiveness for supporting autonomous and online changes of the parallel pattern compositions. Online self-adaptation is achieved with an online profiler that characterizes the applications, which is combined with a new self-adaptive strategy and a model for smooth transitions on reconfigurations. The solution provides a new abstraction layer that enables application programmers to define non-functional requirements instead of hand-tuning complex configurations. Hence, we contribute with additional abstractions and flexible self-adaptation for responsiveness at run-time. The proposed solution is evaluated with applications having different processing characteristics, workloads, and configurations. The results show that it is possible to provide additional abstractions, flexibility, and responsiveness while achieving performance comparable to the best static configuration executions.

show abstract

Section: Results Summarymentioning

confidence: 99%

Section: Decision-making Strategymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Online and transparent self-adaptation of stream parallel patterns

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…Most of them are data-centric using static template instantiation or dynamic runtime polymorphism to model data processing in a pipeline. To name a few popular examples: oneTBB [1] and TPL [20] require explicit definitions of input and output types for each stage; GrPPI [8] provides a composable abstraction for data-and streamparallel patterns with a pluggable back-end to support task scheduling; FastFlow [4] models parallel dataflow using predefined sequential and parallel building blocks; TTG [7] focuses on dataflow programming using various template optimization techniques; SPar [10][11][12]23] analyzes annotated attributes extracted from the data and stream parallelism domain, and automatically generates parallel patterns defined in FastFlow; Proteas [24] introduces a programming model for directive-based parallelization of linear pipeline; [34,35] propose self-adaptive mechanism to decide the degree of parallelism and generate the pattern compositions in FastFlow. These programming models, however, constrain users to design pipeline algorithms using their data models, making it difficult to use especially for applications that only need pipeline scheduling atop custom data structures.…”

Section: Related Workmentioning

confidence: 99%

Pipeflow: An Efficient Task-Parallel Pipeline Programming Framework using Modern C++

Chiu¹,

Huang²,

Guo³

et al. 2022

Preprint

View full text Add to dashboard Cite

Pipeline is a fundamental parallel programming pattern. Mainstream pipeline programming frameworks count on data abstractions to perform pipeline scheduling. This design is convenient for data-centric pipeline applications but inefficient for algorithms that only exploit task parallelism in pipeline. As a result, we introduce a new task-parallel pipeline programming framework called Pipeflow. Pipeflow does not design yet another data abstraction but focuses on the pipeline scheduling itself, enabling more efficient implementation of task-parallel pipeline algorithms than existing frameworks. We have evaluated Pipeflow on both micro-benchmarks and real-world applications. As an example, Pipeflow outperforms oneTBB 24% and 10% faster in a VLSI placement and a timing analysis workloads that adopt pipeline parallelism to speed up runtimes, respectively.

show abstract

“…Self-adaptiveness can also be used for providing additional parallelism abstractions to application programmers, which is a potential opportunity to simplify the process of running stream processing applications. 15,16 However, it is an open question to what extent self-adaptiveness is being applied and how efficient it is for providing parallelism abstractions. This work provides a systematic literature review (SLR) of self-adaptive approaches used in the stream processing scenario.…”

mentioning

confidence: 99%

Self‐adaptation on parallel stream processing: A systematic review

Vogel

Griebler

Danelutto

et al. 2021

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

A recurrent challenge in real-world applications is autonomous management of the executions at run-time. In this vein, stream processing is a class of applications that compute data flowing in the form of streams (e.g., video feeds, images, and data analytics), where parallel computing can help accelerate the executions. On the one hand, stream processing applications are becoming more complex, dynamic, and long-running. On the other hand, it is unfeasible for humans to monitor and manually change the executions continuously. Hence, self-adaptation can reduce costs and human efforts by providing a higher-level abstraction with an autonomic/seamless management of executions. In this work, we aim at providing a literature review regarding self-adaptation applied to the parallel stream processing domain. We present a comprehensive revision using a systematic literature review method. Moreover, we propose a taxonomy to categorize and classify the existing self-adaptive approaches.Finally, applying the taxonomy made it possible to characterize the state-of-the-art, identify trends, and discuss open research challenges and future opportunities.

show abstract

Providing high‐level self‐adaptive abstractions for stream parallelism on multicores

Cited by 11 publications

References 40 publications

Online and transparent self-adaptation of stream parallel patterns

Online and transparent self-adaptation of stream parallel patterns

Pipeflow: An Efficient Task-Parallel Pipeline Programming Framework using Modern C++

Self‐adaptation on parallel stream processing: A systematic review

Contact Info

Product

Resources

About