Filters








550 Hits in 4.1 sec

Oracle-guided scheduling for controlling granularity in implicitly parallel languages

UMUT A. ACAR, ARTHUR CHARGUÉRAUD, MIKE RAINEY
2016 Journal of functional programming  
In this paper, we propose techniques for automatically controlling granularity in implicitly parallel programming languages to achieve parallel efficiency and performance.  ...  This "granularity problem" is especially important in implicitly parallel languages, where the programmer expresses all potential for parallelism, leaving it to the system to exploit parallelism by creating  ...  In this paper, we propose techniques for automatically controlling granularity in implicitly parallel programming languages to achieve parallel efficiency and performance.  ... 
doi:10.1017/s0956796816000101 fatcat:tkb7r7omkfd4biuin5tk4v6gre

Provably and practically efficient granularity control

Umut A. Acar, Vitaly Aksenov, Arthur Charguéraud, Mike Rainey
2019 Proceedings of the 24th Symposium on Principles and Practice of Parallel Programming - PPoPP '19  
In this paper, we present provably and practically efficient techniques for controlling granularity within the run-time system of the language.  ...  Our starting point is "oracle-guided scheduling", a result from the functional-programming community that shows that granularity can be controlled by an "oracle" that can predict the execution time of  ...  As in oracle-guided scheduling [Acar et al. 2016 ], we ask the programmer to provide an abstract cost function, e.g., asymptotic cost, for each piece of parallel code.  ... 
doi:10.1145/3293883.3295725 dblp:conf/ppopp/AcarACR19 fatcat:3lrzgnyfwfepxixn3q2kh7gaqu

Hardware-modulated parallelism in chip multiprocessors

Julia Chen, Philo Juang, Kevin Ko, Gilberto Contreras, David Penry, Ram Rangan, Adam Stoler, Li-Shiuan Peh, Margaret Martonosi
2005 SIGARCH Computer Architecture News  
The difficulties faced in static scheduling were shown in our simulations with a static scheduling algorithm, fed with oracle profiled inputs suffering up to 107% slowdown compared to NDP's hardware scheduler  ...  The software layer is encouraged to expose large amounts of multi-granular, heterogeneous parallelism.  ...  We also wish to thank Niraj Jha, Fei Sun and Anish Muttreja of Princeton for their pointers towards the MCP static scheduling algorithm as a comparison yardstick for NDP and performance macromodels.  ... 
doi:10.1145/1105734.1105742 fatcat:d5iferqj5fghrmndyw6b4ill6y

Tuning the victim selection policy of Intel TBB

Alexandru C. Iordan, Magnus Jahre, Lasse Natvig
2015 Journal of systems architecture  
Parallelization overheads can account for sub-linear speedups and can increase the energy consumption of applications.  ...  Our results showed that failed steals account for the largest overhead. In this work, we focus on TBB's victim selection policy.  ...  The library allows parallelism to be annotated both explicitly and implicitly.  ... 
doi:10.1016/j.sysarc.2015.07.004 fatcat:pwn46qgwhbc53bowdgglegmmhu

A language extension set to generate adaptive versions automatically

Maxime Schmitt, Cédric Bastoul, Philippe Helluy, A. Anciaux-Sedrakian, Q. H. Tran
2018 Oil & Gas Science and Technology  
In this paper we investigate language extensions and compiler tools to achieve that task semi-automatically in the context of approximate computing.  ...  We identified the semantic and syntactic information necessary for a compiler to automatically handle approximation and adaptive techniques for a particular class of programs.  ...  This allows for a code with low control overhead, e:g., the second schedule present in Figure 10 .  ... 
doi:10.2516/ogst/2018049 fatcat:ahnab4jepvalvjw4vtn6qvjjam

How much parallelism is there in irregular applications?

Milind Kulkarni, Martin Burtscher, Rajeshkar Inkulu, Keshav Pingali, Calin Casçaval
2008 Proceedings of the 14th ACM SIGPLAN symposium on Principles and practice of parallel programming - PPoPP '09  
In this paper, we describe the design and implementation of a tool called ParaMeter that produces parallelism profiles for irregular programs.  ...  ParaMeter can also generate constrained parallelism profiles for a fixed number of cores.  ...  We see that the oracle scheduler exhibits much higher parallelism than random scheduling.  ... 
doi:10.1145/1504176.1504181 dblp:conf/ppopp/KulkarniBIPC09 fatcat:5gbsm3qb4neeznvfou4uljjbp4

How much parallelism is there in irregular applications?

Milind Kulkarni, Martin Burtscher, Rajeshkar Inkulu, Keshav Pingali, Calin Casçaval
2009 SIGPLAN notices  
In this paper, we describe the design and implementation of a tool called ParaMeter that produces parallelism profiles for irregular programs.  ...  ParaMeter can also generate constrained parallelism profiles for a fixed number of cores.  ...  We see that the oracle scheduler exhibits much higher parallelism than random scheduling.  ... 
doi:10.1145/1594835.1504181 fatcat:uopajor3jncx5i4delpcyzmrnm

Parallel execution of prolog programs: a survey

Gopal Gupta, Enrico Pontelli, Khayri A.M. Ali, Mats Carlsson, Manuel V. Hermenegildo
2001 ACM Transactions on Programming Languages and Systems  
Since the early days of logic programming, researchers in the field realized the potential for exploitation of parallelism present in the execution of logic programs.  ...  The objective of this article is to provide a comprehensive survey of the issues arising in parallel execution of logic programming languages along with the most relevant approaches explored to date in  ...  A centralized controller is in charge of allocating oracles to idle agents.  ... 
doi:10.1145/504083.504085 fatcat:7dms2hz5mzdsjezpgu2jnf4dsa

Taming the Expressiveness and Programmability of Graph Analytical Queries [article]

Lu Qin, Longbin Lai, Kongzhang Hao, Zhongxin Zhou, Yiwei Zhao, Yuxing Han, Xuemin Lin, Zhengping Qian, Jingren Zhou
2020 arXiv   pre-print
While analyzing existing domain-specific languages (DSLs) for analytical queries regarding the perspectives of completeness, expressiveness and programmability, we find out that none of existing work has  ...  Graph database has enjoyed a boom in the last decade, and graph queries accordingly gain a lot of attentions from both the academia and industry. We focus on analytical queries in this paper.  ...  Graph computation models and primitives are more often developed for large graph processing in the parallel/distributed contexts.  ... 
arXiv:2004.09045v2 fatcat:fdssnpjnvrgkpemze5m7w7bvtu

Analysis and Optimization of Task Granularity on the Java Virtual Machine

Andrea Rosà, Eduardo Rosales, Walter Binder
2019 ACM Transactions on Programming Languages and Systems  
Task granularity, i.e., the amount of work performed by parallel tasks, is a key performance attribute of parallel applications.  ...  We focus on task-parallel applications running in a single Java Virtual Machine on a shared-memory multicore.  ...  ., if the overhead of creating and scheduling the task is higher than the expected benefits of executing work in parallel). Acar et al. [1] propose oracle scheduling.  ... 
doi:10.1145/3338497 fatcat:5t6yjwohjfflfa4nmuvek2di4a

Improving Multi-Document Summarization through Referenced Flexible Extraction with Credit-Awareness [article]

Yun-Zhu Song and Yi-Syuan Chen and Hong-Han Shuai
2022 arXiv   pre-print
To this end, we propose a loss weighting mechanism that makes the model aware of the unequal importance for the sentences not in the pseudo extraction oracle, and leverage the fine-tuned abstractor to  ...  Specifically, we leverage pre-trained language models to construct a hierarchical extractor for salient sentence selection across documents and an abstractor for rewriting the selected contents as summaries  ...  We are grateful to the National Center for High-performance Computing for computer time and facilities.  ... 
arXiv:2205.01889v1 fatcat:uebo7nv56ncfbnyctr3hra5f5e

Limits of instruction-level parallelism

David W. Wall
1991 Proceedings of the fourth international conference on Architectural support for programming languages and operating systems - ASPLOS-IV  
We do work in the design, fabrication and packaging of hardware; language processing and scaling issues in system software design; and the exploration of new applications areas that are opening up with  ...  Such an examination is complicated by the wide variety of hardware and software techniques for increasing the parallelism that can be exploited, including branch prediction, register renaming, and alias  ...  A greedy scheduling algorithm, guided by a configurable oracle, packs these instructions into a sequence of pending cycles.  ... 
doi:10.1145/106972.106991 dblp:conf/asplos/Wall91 fatcat:3izjcs3phjbjzfi7fnefpe3ko4

Limits of instruction-level parallelism

David W. Wall
1991 SIGARCH Computer Architecture News  
We do work in the design, fabrication and packaging of hardware; language processing and scaling issues in system software design; and the exploration of new applications areas that are opening up with  ...  Such an examination is complicated by the wide variety of hardware and software techniques for increasing the parallelism that can be exploited, including branch prediction, register renaming, and alias  ...  A greedy scheduling algorithm, guided by a configurable oracle, packs these instructions into a sequence of pending cycles.  ... 
doi:10.1145/106975.106991 fatcat:gcbwvkqjwna6pk2l2copi4oinm

Limits of instruction-level parallelism

David W. Wall
1991 ACM SIGOPS Operating Systems Review  
We do work in the design, fabrication and packaging of hardware; language processing and scaling issues in system software design; and the exploration of new applications areas that are opening up with  ...  Such an examination is complicated by the wide variety of hardware and software techniques for increasing the parallelism that can be exploited, including branch prediction, register renaming, and alias  ...  A greedy scheduling algorithm, guided by a configurable oracle, packs these instructions into a sequence of pending cycles.  ... 
doi:10.1145/106974.106991 fatcat:n4frf6asgzezlgxvxf5p4d3r3e

Limits of instruction-level parallelism

David W. Wall
1991 SIGPLAN notices  
We do work in the design, fabrication and packaging of hardware; language processing and scaling issues in system software design; and the exploration of new applications areas that are opening up with  ...  Such an examination is complicated by the wide variety of hardware and software techniques for increasing the parallelism that can be exploited, including branch prediction, register renaming, and alias  ...  A greedy scheduling algorithm, guided by a configurable oracle, packs these instructions into a sequence of pending cycles.  ... 
doi:10.1145/106973.106991 fatcat:oeclvb2h5fegfo73jatt6oi4pe
« Previous Showing results 1 — 15 out of 550 results