Phasers: a unified deadlock-free construct for collective and point-to-point synchronization
The DOI matches its Crossref record; 12 of the 12 references it cites were found in a registry or an index.
Abstract
Coordination and synchronization of parallel tasks is a major source of complexity in parallel programming. These constructs take many forms in practice including mutual exclusion in accesses to shared resources, termination detection of child tasks, collective barrier synchronization, and point-to-point synchronization. In this paper, we introduce phasers, a new coordination construct that unifies collective and point-to-point synchronizations. We establish two safety properties for phasers: deadlock-freedom and phase-ordering. Performance results obtained from a portable implementation of phasers on three different SMP platforms demonstrate that phasers can deliver superior performance to existing barrier implementations, in addition to the productivity benefits that result from their generality and safety properties.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“In this example of manycore processor speedup, a fat core needing 10 times as many resources would be more effective than the 10 thin cores it replaces.5,15”
From A view of the parallel computing landscape (2009) · cites this paper as [15]
Citations
Cites
Transactional memory: architectural support for lock-free data structures
Herlihy, M., & Moss, J. E. B. (1993) · Proceedings of the 20th annual international symposium on Computer architecture - ISCA '93 · doi:10.1145/165123.165164 · Cited by 2,141
X10: an object-oriented approach to non-uniform cluster computing
Charles, P., Grothoff, C., Saraswat, V., Donawa, C., Kielstra, A., Ebcioglu, K., von Praun, C., & Sarkar, V. (2005) · Proceedings of the 20th annual ACM SIGPLAN conference on Object-oriented programming, systems, languages, and applications · doi:10.1145/1094811.1094852 · Cited by 1,268 · cited 1×
“One of the major obstacles to improved productivity is the complexity of coordination and synchronization of parallel tasks that is inherent in current parallel programming models.”
BackgroundSelf-citationCilk: an efficient multithreaded runtime system
Blumofe, R. D., Joerg, C. F., Kuszmaul, B. C., Leiserson, C. E., Randall, K. H., & Zhou, Y. (1995) · Proceedings of the fifth ACM SIGPLAN symposium on Principles and practice of parallel programming · doi:10.1145/209936.209958 · Cited by 1,021 · cited 2×
“parallel programming include transactional memory systems for mutual exclusion [7], work stealing schedulers to support dynamic nested parallel execution model with lightweight task creation and termination detection as in Cilk [11], and deadlock-free barrier synchronizations as in X10’s clocks [2].”
BackgroundInfluentialPeierls, T., Goetz, B., Bloch, J., Bowbeer, J., Lea, D., & Holmes, D. (2006) · Medical Entomology and Zoology · Cited by 424
Goetz, B., Peierls, T., Bloch, J., Bowbeer, J., Holmes, D., & Lea, D. (2006) · Cited by 264
Productivity and performance using partitioned global address space languages
Yelick, K., Bonachea, D., Chen, W.-Y., Colella, P., Datta, K., Duell, J., Graham, S. L., Hargrove, P., Hilfinger, P., Husbands, P., et al. (2007) · Proceedings of the 2007 international workshop on Parallel symbolic computation · doi:10.1145/1278177.1278183 · Cited by 192 · cited 2×
“These constraints are imposed to ensure the integrity of the single statement [16].”
BackgroundThe fuzzy barrier: a mechanism for high speed synchronization of processors
Gupta, R. (1989) · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) · doi:10.1145/70082.68187 · Cited by 113 · cited 3×
“Gupta’s work on fuzzy barriers [5] introduces the concept of overlapping synchronization with “real” work in a model that is now widely known as a split-phase barrier.”
BackgroundTitanium Language Reference Manual
Hilfinger, P. N., Bonachea, D., Graham, S. L., Liblit, B., Pike, G., & Yelick, K. (2001) · Cited by 94
The fuzzy barrier: a mechanism for high speed synchronization of processors
Gupta, R. (1989) · ACM SIGARCH Computer Architecture News · doi:10.1145/68182.68187 · Cited by 92
Deadlock-free scheduling of X10 computations with bounded resources
Agarwal, S., Barik, R., Bonachea, D., Sarkar, V., Shyamasundar, R. K., & Yelick, K. (2007) · ACM Symposium on Parallelism in Algorithms and Architectures (SPAA) · doi:10.1145/1248377.1248416 · Cited by 57 · cited 3×
“In this section we precisely define the guarantees phasers provide for phase ordering and deadlock avoidance by building on the the notion of an X10 computation DAG introduced in [1] for finish and async constructs.”
BackgroundMethodInfluentialSelf-citation
Research works printed but not linked: 1
- L. A. Smith and J. M. Bull . A multithreaded java grande benchmark suite . In Proceedings of the Third Workshop on Java for High Performance Computing , June 2001 . L. A. Smith and J. M. Bull. A multithreaded java grande benchmark suite. In Proceedings of the Third Workshop on Java for High Performance Computing, June 2001.
Cited by
A view of the parallel computing landscape
Asanovic, K., Bodik, R., Demmel, J., Keaveny, T., Keutzer, K., Kubiatowicz, J., Morgan, N., Patterson, D., Sen, K., Wawrzynek, J., et al. (2009) · Communications of the ACM · doi:10.1145/1562764.1562783 · Cited by 627
NVIDIA A100 Tensor Core GPU: Performance and Innovation
Choquette, J., Gandhi, W., Giroux, O., Stam, N., & Krashinsky, R. (2021) · IEEE Micro · doi:10.1109/mm.2021.3061394 · Cited by 410
Habanero-Java: the new adventures of old X10
Cavé, V., Zhao, J., Shirako, J., & Sarkar, V. (2011) · Principles and Practice of Programming in Java · doi:10.1145/2093157.2093165 · Cited by 263
“result() is only computed once per loop iteration since it is included in a single statement [33].”
BackgroundMethodInfluentialSelf-citationWork-first and help-first scheduling policies for async-finish task parallelism
Guo, Y., Barik, R., Raman, R., & Sarkar, V. (2009) · IEEE International Parallel and Distributed Processing Symposium (IPDPS) · doi:10.1109/ipdps.2009.5161079 · Cited by 180
“To that end, we focused on X10 benchmarks that have been used in past evaluations [6, 3, 16, 18 , 9].”
BackgroundMethodSelf-citationIntegrating Asynchronous Task Parallelism with MPI
Chatterjee, S., Taşırlar, S., Budimlić, Z., Cavé, V., Chabbi, M., Grossman, M., Sarkar, V., & Yan, Y. (2013) · IEEE International Parallel and Distributed Processing Symposium (IPDPS) · doi:10.1109/ipdps.2013.78 · Cited by 106
“HC supports phasers [12], [13] for fine-grained synchronization among dynamically created tasks.”
BackgroundSelf-citation
Metadata
- Authors
- Shirako, J., Peixotto, D. M., Sarkar, V., & Scherer, W. N.
- Affiliations
- Rice University (US)
- Venue
- Proceedings of the 22nd annual international conference on Supercomputing
- Publisher
- ACM
- Published
- 2008-06-07
- Pages
- 277-288
- OpenAlex
- W1970787534
- Cited by
- 124
- References
- 12
- Citations by year
- 2023: 1 · 2021: 2 · 2020: 1 · 2019: 3 · 2018: 2 · 2017: 8 · 2016: 3 · 2015: 13 · 2014: 12 · 2013: 18 · 2012: 12
- FWCI
- 9.61
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- International Conference on Supercomputing (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Hardware and Architecture › Parallel Computing and Optimization Techniques
- Physical Sciences › Computer Science › Computer Networks and Communications › Distributed systems and fault tolerance
- Physical Sciences › Computer Science › Software › Logic, programming, and type systems
- Keywords
- Phaser, barrier synchronization, collective synchronization, deadlock freedom, phase ordering, shared resource access, parallel programming
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Phasers: a unified deadlock-free construct for collective and point-to-point synchronization. https://app.cochat.ai/papers/10.1145/1375527.1375568
- The paper
- Shirako, J., Peixotto, D. M., Sarkar, V., & Scherer, W. N. (2008). Phasers: a unified deadlock-free construct for collective and point-to-point synchronization. Proceedings of the 22nd annual international conference on Supercomputing. https://doi.org/10.1145/1375527.1375568
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.