Locality Aware Task Scheduling in Parallel Data Stream Processing

Kruliš, Martin; Bednárek, David; Yaghob, Jakub; Zavoral, Filip

doi:10.1007/978-3-319-10422-5_35

Cited by 10 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bobox is a framework [1], [2], [6] for task-based parallel programming [4]. The framework uses a number of worker threads, each one with its own cooperative scheduler.…”

Section: B Bobox Frameworkmentioning

confidence: 99%

See 1 more Smart Citation

Using static code analysis to improve coarse task granularity in bobox

Krizik

Zavoral

Čermák

2015

2015 Tenth International Conference on Digital Information Management (ICDIM)

Self Cite

View full text Add to dashboard Cite

Long-running tasks inhibit a parallel ex ecution in some cases. A finer task granularity can significantly improve execution times in parallel en vironment. That is even more significant when using cooperative scheduling and it puts higher requirements on user code. In this paper we present a method of user code optimization using static code analysis in order to reduce long-running tasks. The method is based on evaluation of runtime code complexity and yielding a task execution in an appropriate place.

show abstract

“…Bobox is a framework [1], [2], [6] for task-based parallel programming [4]. The framework uses a number of worker threads, each one with its own cooperative scheduler.…”

Section: B Bobox Frameworkmentioning

confidence: 99%

“…The Bobox scheduler [5] is a cooperative scheduler used in the Bobox framework [1], thus applications based on it inherit its behavior (see Sec. I-B).…”

Section: Introductionmentioning

confidence: 99%

Using static code analysis to improve coarse task granularity in bobox

Krizik

Zavoral

Čermák

2015

2015 Tenth International Conference on Digital Information Management (ICDIM)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Similarly, there are many efforts that introduce methods whose focus is to improve the metrics across multiple queries [51,52,53,54,55,56,57,54,58,59,60,61]. [51] explores the multi-query scheduling as a job-scheduling problem and utilizes real-time computing approaches.…”

Section: Related Workmentioning

confidence: 99%

“…[51] explores the multi-query scheduling as a job-scheduling problem and utilizes real-time computing approaches. [52] explored task schedulers that consider the hardware statistics in parallel processing systems that execute multiple streaming queries. [53] proposes an adaption to chain scheduling strategy to minimize the memory overhead and output latency.…”

Section: Related Workmentioning

confidence: 99%

Practical Identification of Dynamic Precedence Criteria to Produce Critical Results from Big Data Streams

Works

Rundensteiner

2015

Big Data Research

View full text Add to dashboard Cite

“…Non-Uniform Memory Access (NUMA) systems overcome the drawbacks of the Uniform Memory Access (UMA) systems such as lack of performance with increasing the number of processing units and need for synchronization barriers to ensure the correctness of the shared memory accesses [1]. NUMA systems have a memory subsystem for each CPU, considering that the CPU is placed on the last level in the tree-based memory hierarchy [2]. While the number of processing units is continuously growing, NUMA systems avoid congested data traffic, which is one of the main factors that influence performance.…”

Section: Introductionmentioning

confidence: 99%

Extending NUMA-BTLP Algorithm with Thread Mapping Based on a Communication Tree

Știrb

2018

Computers

View full text Add to dashboard Cite

The paper presents a Non-Uniform Memory Access (NUMA)-aware compiler optimization for task-level parallel code. The optimization is based on Non-Uniform Memory Access—Balanced Task and Loop Parallelism (NUMA-BTLP) algorithm Ştirb, 2018. The algorithm gets the type of each thread in the source code based on a static analysis of the code. After assigning a type to each thread, NUMA-BTLP Ştirb, 2018 calls NUMA-BTDM mapping algorithm Ştirb, 2016 which uses PThreads routine pthread_setaffinity_np to set the CPU affinities of the threads (i.e., thread-to-core associations) based on their type. The algorithms perform an improve thread mapping for NUMA systems by mapping threads that share data on the same core(s), allowing fast access to L1 cache data. The paper proves that PThreads based task-level parallel code which is optimized by NUMA-BTLP Ştirb, 2018 and NUMA-BTDM Ştirb, 2016 at compile-time, is running time and energy efficiently on NUMA systems. The results show that the energy is optimized with up to 5% at the same execution time for one of the tested real benchmarks and up to 15% for another benchmark running in infinite loop. The algorithms can be used on real-time control systems such as client/server based applications which require efficient access to shared resources. Most often, task parallelism is used in the implementation of the server and loop parallelism is used for the client.

show abstract

Locality Aware Task Scheduling in Parallel Data Stream Processing

Cited by 10 publications

References 14 publications

Using static code analysis to improve coarse task granularity in bobox

Using static code analysis to improve coarse task granularity in bobox

Practical Identification of Dynamic Precedence Criteria to Produce Critical Results from Big Data Streams

Extending NUMA-BTLP Algorithm with Thread Mapping Based on a Communication Tree

Contact Info

Product

Resources

About