Erik Saule's Homepage

Research

An overview of what I do as a word cloud can be found here

Generalities

I am currently holding an Associate Professor position in the University of North Carolina at Charlotte in the department of Computer Science.

I used to be a post-doctoral researcher position at The Ohio State University in the BioMedical Informatics in Umit Catalyurek's High Performance Computing group.

I did my PhD in Computer Science under the supervision of Denis Trystram as part of the MOAIS team at LIG laboratory in Grenoble (France). My PhD manuscript is entitled "Approximation Algorithms for Multi-Objective Scheduling. Application to Parallel and Embedded Systems" and is available (in french).

I frequently serve in the organization and program committee of various conferences such as IPDPS, SC, HiPC, ICPP, See a complete list here.

I am frequently reviewing papers for the following journals : Journal of Scheduling, Parallel Computing, Journal of Supercomputing, IEEE Transactions on Parallel and Distributed Systems, Parallel Computing and Journal of Parallel and Distributed Computing.

I am a co guest editor for Parallel Computing special issue on heterogeneity.

Projects available

I am interested in advising students (undergraduate and graduate) on various projects. Check out the available projects for details.

Research Interest

My research is focused on combinatorial optimization problems for high performance computing.

Theoretically guaranteed heuristics

Most relevant optimization problems in high performance computing are NP-Hard. It is unlikely that a polynomial optimal algorithm can be designed. However, it does not mean that reasonable solutions can not be found. Approximation algorithms are a class of polynomial heuristics which produce solutions whose objective values are guaranteed to be close to the optimal value (either in relative or absolute terms). Approximation algorithms measure the quality of a solution by measuring the degradation in term of objective value. Resource augmentation algorithms measure the quality of a solution in term of amount of extra resource that need to be allocated to compensate for the non optimality of the solution.

See for instance: [P1], [J18], [C69], [C58], [C61], [C66], [C65], [C68].

Multi objective optimization

When systems got more complex, the notion of performance becomes less clearly defined. It is therefore not sufficient to optimize a single performance index. For instance, when computing systems become larger, the appearance of system failure is more likely. Therefore, the performance of the system must include reliability consideration such as probability of failure or number of fault the system can tolerate. Another example is the importance of taking the energy into account to increase battery life of mobile device.

When optimizing more than one objective function, optimality is no longer well defined. The notion Pareto dominance defines a partial order between the solutions. Pareto optimal solutions are the set of solution that can not be improved on one objective without degrading another one.

See for instance: [P1], [B2], [J18], [J19], [C61], [C66], [C65], [C68].

Spatial partitioning

In many distributed applications, the computations take place in a discrete 2D or 3D space. At each iteration of the application, the state of each cell is updated using the state of neighbor cells in a stencil fashion. In applications such as particle-in-cell simulation and raycasting, the computation time required to update each cell may vary in function of well known parameters (such as the distribution of the particle in the field or the position of the objects in the scene). Spatial partitioning focuses on distributing such applications on a parallel computer. The cells must be allocated to the processor meticulously by taking into account both the load balance between the processors and the communication time between two adjacent cells. Methods should also take into account the possible dynamic aspects of such systems. To solve this problem, we are proposing new rectangular decomposition algorithms and analyze their behavior both theoretically and practically.

See for instance: [J14] [C55], [R10], [T14], [T16].

Cluster Scheduling

Clusters are a common platform in high performance computing. It is frequent that more than one application run at a given time on a single cluster. From that fact arise the problem of allocating a slice of the cluster to each application. Cluster scheduling is the field that establishes models to differentiate between possible allocations and provided algorithms to solve them. Hot topics in cluster scheduling include moldable tasks, multi user scheduling, fairness, stretch and reservation.

See for instance: [P1], [C69], [C58], [C61], [C66], [C65].

Application Scheduling

When executing an application on a parallel system, one must decide on which execution unit each computation will occur. Application scheduling handles that question by proposing different model for the application and for the platform. A common way to model applications is the DAG scheduling model. Computations are grouped into tasks and precedence dependencies are given as a graph. Provided computation time for the tasks and communication time for each precedence dependency, the problem is to map the tasks on the computing units to minimize the application execution time. Other models consider applications that handles flows of data. The execution is then consider in its steady state and the goal is to maximize the application throughput.

See for instance: [P1], [J18], [J19], [C69], [C58], [C66], [C68], [R11].

Large Scale Graph Mining

Graphs are a common representation of complex data. They are commonly used to represent social interactions, scientific citation, traffic patterns, consumer choices, sparse linear algebra computations, etc.. With the growth of the Internet, the amount of data available grew exponentially in all domains. Facebook has more than 1 billion users. Amazon has hundreds of million of customers and products. Millions of scientific publications have been written. I am interested in developing new models and algorithms to enable the analysis of very large graphs. I am currently interested in in analysis such as link prediction, recommendation, popularity measures and community detections.

See for instance: [R8], [C48], [C50], [R7], [C42], [C44]

High Performance Graph Algorithms and Sparse Matrix Computations

The size of the graphs and the complexity of the analysis required on them make the study of algorithms on large graphs relevant both in practice and as a case study. Because sparse matrices and graphs are algrebraically similar, most of the techniques that apply on one also apply on the other one. I am interested in optimizing graph kernels and sparse matrices computation using high performance computing techniques such as parallel computing, distributed computing, accelerators but also in algorithm engineering.

See for instance: [C38], [C47], [C54], [C51], [C52], [], [C44]

Middleware

It is no doubt that, with enough work, High Performance Computing experts will achieve most the available performance of a system toward a particular application. To enable non-HPC experts to utilize computing resources to their maximum, tools such as middleware and programming models were developed. Each tool targets a particular class of platform and applications and with different concerns in mind. I believe it is of key importance to study the properties of middleware and programming models to identify their strengths and weaknesses.

See for instance: [J15], [C57], [C62], [C49], [C46]

Publications

Here is a list of my publications:

PhD

[P1]
Erik Saule. Algorithmes d'approximation pour l'ordonnancement multi-objectif. Applications aux systèmes parallèles et embarqués. PhD thesis, Institut polytechnique de Grenoble, 2008. [ bib | .pdf ]

Book Chapters

[B1]
Erik Saule, Hasan Metin Aktulga, Chao Yang, Esmond G. Ng, and Ümit V. Çatalyürek. An Out-of-Core Task-based Middleware for Data-Intensive Scientific Computing. Handbook on Data Centers. Springer, 2015. ISBN 978-1-4939-2091-4. [ bib | DOI ]
[B2]
Pierre-Francois Dutot, Krzysztof Rzadca, Erik Saule, and Denis Trystram. Multi-objective scheduling, chapter 9. Introduction to scheduling. Chapman and Hall/CRC Press, November 2009. ISBN: 978-1420072730. [ bib ]
[B3]
Xavier Besseron, Slim Bouguerra, Thierry Gautier, Erik Saule, and Denis Trystram. Fault tolerance and availability awarness in computational grids, chapter 5. Fundamentals of Grid Computing. Chapman and Hall/CRC Press, December 2009. ISBN: 978-1439803677. [ bib ]

Journals

[J1]
David Burlinson, Matthew Mcquaigue, Alec Goncharow, Kalpathi Subramanian, Erik Saule, Jamie Payton, and Paula Goolkasian. Bridges: Real world data, assignments and visualizations to engage and motivate cs majors. Education and Information Technologies, 2023. [ bib | DOI | .pdf ]
[J2]
Alec Goncharow, Matthew Mcquaigue, Erik Saule, Kalpathi Subramanian, Paula Goolkasian, and Jamie Payton. Cs-materials: A system for classifying and analyzing pedagogical materials to improve adoption of parallel and distributed computing topics in early cs courses. Journal of Parallel and Distributed Computing, 2021. [ bib | DOI | .pdf ]
[J3]
Pourya Naderi Yeganeh, Christine Richardson, Erik Saule, Ann Loraine, and M. Taghi Mostafavi. Revisiting the use of graph centrality models in biological pathway analysis. BioData Mining, 2020. [ bib | DOI | .pdf ]
[J4]
Ahmet Erdem Sariyüce, Kamer Kaya, Erik Saule, and Ümit V. Çatalyürek. Graph manipulations for fast centrality computation. ACM Transactions on Knowledge Discovery from Data (TKDD), 11, 2017. [ bib | DOI | .pdf ]
[J5]
T. Dytrych, P. Maris, K. D. Launey, J. P. Draayer, J. P. Vary, D. Langr, E. Saule, M. A. Caprio, Ü. Çatalyürek, and M. Sosonkina. Efficacy of the su(3) scheme for ab initio large-scale calculations beyond the lightest nuclei. Computer Physics Communications, 207:202--210, October 2016. [ bib | DOI ]
[J6]
Ahmet Erdem Sariyüce, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Incremental closeness centrality in distributed memory. Parallel Computing, 47:3--18, August 2015. [ bib | DOI | http | .pdf ]
[J7]
Ahmet Erdem Sariyüce, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Regularizing graph centrality computations. Journal of Parallel and Distributed Computing, 76:106--119, February 2015. [ bib | DOI | .pdf ]
[J8]
Myoungsoo Jung, Ellis H. Wilson III, Wonil Choi, John Shalf, Hasan Metin Aktulga, Chao Yang, Erik Saule, Ümit V. Çatalyürek, and Mahmut Kandemir. Exploring the future of out-of-core computing with compute-local non-volatile memory. Scientific Programming, 22(2):125--139, 2014. [ bib | DOI | http ]
[J9]
Onur Küçüktunç, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Diversifying citation recommendations. ACM Transactions on Intelligent Systems and Technology, 5(4), 2014. [ bib | .pdf ]
[J10]
T. Dytrych, K. D. Launey, J. P. Draayer, P. Maris, J. P. Vary, E. Saule, Ü. Çatalyürek, M. Sosonkina, D. Langr, and M. A. Caprio. Collective modes in light nuclei from first principles. Physical Review Letters, 111(25), December 2013. [ bib | DOI ]
[J11]
Onur Küçüktunç, Kamer Kaya, Erik Saule, and Ümit V. Çatalyürek. Fast recommendation on bibliographic networks with sparse-matrix ordering and partitioning. Social Network Analysis and Mining, 3(4):1097--1111, December 2013. [ bib | DOI | .pdf ]
[J12]
Anne Benoit, Ümit Çatalyürek, Yves Robert, and Erik Saule. A Survey of Pipelined Workflow Scheduling: Models and Algorithms. ACM Computing Surveys, 45(4), August 2013. [ bib | DOI | .pdf ]
[J13]
Pieter Maris, H Metin Aktulga, Sven Binder, Angelo Calci, Ümit V Çatalyürek, Joachim Langhammer, Esmond Ng, Erik Saule, Robert Roth, James P Vary, and Chao Yang. No core ci calculations for light nuclei with chiral 2- and 3-body forces. Journal of Physics: Conference Series, 454(1):012063, 2013. [ bib | DOI ]
[J14]
Erik Saule, Erdeniz O. Bas, and Umit V. Catalyurek. Load-balancing spatially located computations using rectangular partitions. Journal of Parallel and Distributed Computing, 2012. [ bib | DOI | .pdf ]
[J15]
Timothy D. R. Hartley, Erik Saule, and Umit V. Catalyurek. Improving performance of adaptive component-based dataflow middleware. Parallel Computing, 38(6-7):289--309, 2012. [ bib | DOI | .pdf ]
[J16]
Erik Saule, Doruk Bozdag, and Ümit V. Çatalyürek. Optimizing the maximum stretch of independent tasks on a cluster : From sequential tasks to moldable tasks. Journal of Parallel and Distributed Computing, 72(4):489--503, April 2012. [ bib | DOI | .pdf ]
[J17]
Emmanuel Jeannot, Erik Saule, and Denis Trystram. Optimizing performance and reliability on heterogeneous parallel systems: Approximation algorithms and heuristics. Journal of Parallel and Distributed Computing, 72(2):268 -- 280, February 2012. [ bib | DOI | .pdf ]
[J18]
Erik Saule and Denis Trystram. Analyzing scheduling with transient failures. Information Processing Letters, 109(11):539--542, May 2009. [ bib | DOI | .pdf ]
[J19]
Alain Girault, Erik Saule, and Denis Trystram. Reliability versus performance for critical applications. Journal of Parallel and Distributed Computing, 69(3):326--336, March 2009. [ bib | DOI | .pdf ]

Peer-Review Conferences and Workshops

[C1]
Matthew McQuaigue, Mack Larson, Philip Smith, Sydney Melech, Kalpathi Subramanian, and Erik Saule. Engaging cs1 students with audio themed assignments. In Proc. of CCSC NE, 2024. [ bib | .pdf ]
[C2]
H. Martin Bucker, Jeremiah Corrado, Daniel Fedorin, Diego Garcia-Alvarez, Arturo Gonzalez-Escribano, John Li, Maria Pantoka, Erik Pautsch, Marieke Plesske, Marcelo Ponce, Silvio Rizzi, Erik Saule, Johannes Schoder, George K. Thiruvathukal, Ramses van Zon, Wolf Weber, and David Bunde. Peachy parallel assignments (eduhpc 2023). In Proceedings of SC23 Workshops (SC-W); EduHPC, 2023. [ bib | DOI | slides | .pdf ]
[C3]
Matthew Mcquaigue, Erik Saule, Kalpathi Subramanian, and Jamie Payton. Data-driven discovery of anchor points for pdc content. In Proceedings of SC23 Workshops (SC-W); EduHPC, 2023. [ bib | DOI | slides | .pdf ]
[C4]
Dante Durrman and Erik Saule. Optimizing the critical path of distributed dataflow graph algorithms. In Proceedings of IPDPS Workshops (IPDPSW); PDCO, 2023. [ bib | slides | .pdf ]
[C5]
Md Maruf Hossain and Erik Saule. Postmortem graph analysis on temporal graphs. In Proc of ICPP, August 2022. [ bib | DOI | slides | YouTube | .pdf ]
[C6]
Dante Durrman and Erik Saule. Coloring the vertices of 9-pt and 27-pt stencils with intervals. In Proc. of IPDPS, May 2022. [ bib | slides | YouTube | .pdf ]
[C7]
Katheryn Perry, Cedric Sirianni, Owen Bechtel, Kalpathi Subramanian, and Erik Saule. High school bridges: Visualizations of data, data structures, and more,. In Proc. of SIGCSE, 2022. SIGCSE Demo. [ bib ]
[C8]
Matthew McQuaigue, Jay Strahler, Kalpathi Subramanian, and Erik Saule. Location based assignments in early cs courses using bridges engages students. In Proc. of CCSC SE, 2022. [ bib | .pdf ]
[C9]
Alec Goncharow, Matthew Mcquaigue, Erik Saule, Kalpathi Subramanian, Jamie Payton, and Paula Goolkasian. Mapping materials to curriculum standards for design, alignment, audit, and search. In Proceedings of the 52nd ACM Technical Symposium on Computer Science Education, SIGCSE '21, page 295–301, New York, NY, USA, 2021. Association for Computing Machinery. [ bib | DOI | YouTube | .pdf ]
[C10]
Erik Saule, Kalpathi Subramanian, and Jamie Payton. We need community effort to achieve pdc adoption! In IEEE 28th International Conference on High Performance Computing, Data and Analytics Workshop (HiPCW), 2021. (invited) (EduHiPC). [ bib | DOI | YouTube | .pdf ]
[C11]
Md Maruf Hossain and Erik Saule. Impact of avx-512 instructions on graph partitioning problems. In proc. of P2S2, 2021. [ bib | DOI | slides | YouTube | .pdf ]
[C12]
Kathryn Perry, Kalpathi Subramanian, and Erik Saule. Some bridges span more than water: Engaging high school java learners with data structure visualizations and real-world data. In Proceedings of the 52nd ACM Technical Symposium on Computer Science Education, SIGCSE '21, page 1371, New York, NY, USA, 2021. Association for Computing Machinery. SIGCSE Lightning talk. [ bib | DOI ]
[C13]
Jason Strahler, Matthew Mcquaigue, Alec Goncharow, David Burlinson, Kalpathi Subramanian, Erik Saule, and Jamie Payton. Real-world assignments at scale to reinforce the importance of algorithms and complexity. In Proc. CCSC NE, 2020. [ bib | YouTube | .pdf ]
[C14]
Allie Beckman, Matthew Mcquaigue, Alec Goncharow, David Burlinson, Kalpathi Subramanian, Erik Saule, and Jamie Payton. Engaging early programming students with modern assignments using bridges. In Proc. CCSC CP, 2020. [ bib | .pdf ]
[C15]
Matthew Mcquaigue, Allie Beckman, David Burlinson, Luke Sloop, Alec Goncharow, Erik Saule, Kalpathi Subramanian, and Jamie Payton. An engaging cs1 curriculum using bridges. In Proceedings of the 51st ACM Technical Symposium on Computer Science Education, SIGCSE '20, page 1317, New York, NY, USA, 2020. Association for Computing Machinery. SIGCSE Poster. [ bib | DOI | poster ]
[C16]
David Burlinson, Erik Saule, and Kalpathi Subramanian. Building simple games with bridges. In Proceedings of the 50th ACM Technical Symposium on Computer Science Education, SIGCSE '19, pages 1288--1288, New York, NY, USA, 2019. ACM. SIGCSE Demo. [ bib | DOI ]
[C17]
Alec Goncharow, Anna Boekelheide, Matthew Mcquaigue, David Burlinson, Erik Saule, Kalpathi Subramanian, and Jamie Payton. Classifying pedagogical material to improve adoption of parallel and distributed computing topics. In Proc. of IPDPSW 2019, May 2019. [ bib | .pdf ]
[C18]
Pourya Naderi Yeganeh, Erik Saule, and M. Taghi Mostafavi. Centrality of cancer-related genes in human biological pathways: A graph analysis perspective. In Proc. of BIBM, December 2018. [ bib | DOI | .pdf ]
[C19]
Haofeng Jia and Erik Saule. Local is good: A fast citation recommendation approach. In Proc of ECIR, 2018. acceptance rate: 35%. [ bib | DOI | poster | .pdf ]
[C20]
Matthew Mcquaigue, David Burlinson, Kalpathi Subramanian, Erik Saule, and Jamie Payton. Integrating visualization, assessment and analytics in data structures learning modules. In Proc of SIGCSE, 2018. [ bib | DOI | .pdf ]
[C21]
Haofeng Jia and Erik Saule. Addressing overgeneration error: An effective and efficient approach to keyphrase extraction from scientific papers. In Proc. of BIRNDL 2018, July 2018. [ bib | poster | .pdf ]
[C22]
Erik Saule. Experiences on teaching parallel and distributed computing for undergraduates. In Proc of IPDPSW 2018, May 2018. EduPar best paper. [ bib | .pdf ]
[C23]
Haofeng Jia and Erik Saule. An analysis of citation recommender systems: Beyond the obvious. In Proc of ASONAM, 2017. acceptance rate: 17.2%. [ bib | DOI | .pdf ]
[C24]
Mustafa Kemal Taş, Kamer Kaya, and Erik Saule. Greed is good: Optimistic algorithms for bipartite-graph partial coloring on multicore architectures. In Proc of ICPP 2017, 2017. acceptance rate: 28.4%. [ bib | DOI | .pdf ]
[C25]
Erik Saule, Dinesh Panchananam, Alexander Hohl, Wenwu Tang, and Eric Delmelle. Parallel space-time kernel density estimation. In Proc of ICPP 2017, 2017. acceptance rate: 28.4%. [ bib | DOI | slides | .pdf ]
[C26]
Pierre-Francois Dutot, Erik Saule, Abhinav Srivastav, and Denis Trystram. Online non-preemptive scheduling to optimize max stretch on a single machine. In proc of COCOON 2016, August 2016. [ bib ]
[C27]
Pierre-Francois Dutot, Erik Saule, Abhinav Srivastav, and Denis Trystram. Online non-preemptive scheduling to optimize stretch. In 12th Workshop on Models and Algorithms for Planning and Scheduling Problems, June 2015. [ bib ]
[C28]
Nathanaël Cheriere and Erik Saule. Distributed load balancing for fully heterogeneous machines. In 29th International Symposium on Parallel and Distributed Processing, Workshops and PhD Forum (IPDPSW), Workshop on Heterogeneity in Computing Workshop (HCW), May 2015. [ bib | .pdf ]
[C29]
Manmohan Chaubey and Erik Saule. Replicated data placement for uncertain scheduling. In 29th International Symposium on Parallel and Distributed Processing, Workshops and PhD Forum (IPDPSW), Workshop on Advances in Parallel and Distributed Computational Models (APDCM), May 2015. [ bib | .pdf ]
[C30]
Gordon Erlebacher, Erik Saule, Natasha Flyer, and Evan Bollig. Acceleration of derivative calculations with application to radial basis function - finite-differences on the Intel MIC architecture. In Proc. of International Conference on Supercomputing (ICS), 2014. acceptance rate: 20%. [ bib | .pdf ]
[C31]
P. Calyam, A. Berryman, Erik Saule, H. Subramoni, P. Schopis, G. Springer, Ümit V. Çatalyürek, and D.K. Panda. Wide-area overlay networking to manage science DMZ accelerated flows. In Computing, Networking and Communications (ICNC), 2014 International Conference on, pages 269--275, February 2014. [ bib | DOI ]
[C32]
Ahmet Erdem Sariyüce, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Computing the closeness centrality of evolving networks on clusters. In SIAM Workshop on Network Science (NS14), July 2014. [ bib | poster | .pdf ]
[C33]
Ahmet Erdem Sariyüce, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Hardware/software vectorization for closeness centrality on multi-/many-core architectures. In 28th International Symposium on Parallel and Distributed Processing, Workshops and PhD Forum (IPDPSW), Workshop on Multithreaded Architectures and Applications (MTAAP), 2014. [ bib | slides | .pdf ]
[C34]
Ahmet Erdem Sariyüce, Kamer Kaya, Erik Saule, and Ümit V. Çatalyürek. Incremental algorithms for closeness centrality. In Proc. of IEEE BigData 2013, October 2013. acceptance rate: 37%. [ bib | .pdf ]
[C35]
Ahmet Erdem Sariyüce, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Streamer: a distributed framework for incremental closeness centrality computation. In Proc of IEEE Cluster 2013, September 2013. acceptance rate: 31%. [ bib | .pdf ]
[C36]
Anas Abu-Doleh, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Masher: Mapping long(er) reads with hash-based genome indexing on GPUs. In Proc of ACM Conference on Bioinformatics, Computational Biology and Biomedical Informatics (BCB), September 2013. [ bib | slides | .pdf ]
[C37]
Myoungsoo Jung, Ellis H. Wilson III, Wonil Choi, John Shalf, Hasan Metin Aktulga, Chao Yang, Erik Saule, Ümit V. Çatalyürek, and Mahmut Kandemir. Exploring the future of out-of-core computing with compute-local non-volatile memory. In Proc. of Conference on High Performance Computing Networking, Storage and Analysis (SC '13), November 2013. acceptance rate: 20%. [ bib ]
[C38]
Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Performance evaluation of sparse matrix multiplication kernels on Intel Xeon Phi. In Proc of the 10th Int'l Conf. on Parallel Processing and Applied Mathematics (PPAM), page 10, September 2013. [ bib | slides | .pdf ]
[C39]
Onur Küçüktunç, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. TheAdvisor: A webservice for academic recommendation. In ACM/IEEE Joint Conference on Digital Libraries (JCDL 2013), page 2, July 2013. (poster). [ bib | poster | .pdf ]
[C40]
Onur Küçüktunç, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Towards a personalized, scalable, and exploratory academic recommendation service. In IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), August 2013. acceptance rate: 28%. [ bib | .pdf ]
[C41]
Onur Küçüktunç, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Diversified recommendation on graphs: Pitfalls, measures, and algorithms. In 22nd International World Wide Web Conference (WWW), May 2013. acceptance rate: 15%. [ bib | .pdf ]
[C42]
Ahmet Erdem Sariyüce, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Shattering and compressing networks for betweenness centrality. In SIAM International Conference on Data Mining, SDM, May 2013. acceptance rate: 25.5%. [ bib | .pdf ]
[C43]
Onur Küçüktunç, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Result diversification in automatic citation recommendation. In iConference Workshop on Computational Scientometrics: Theory and Applications, page 4, February 2013. [ bib | .pdf ]
[C44]
Ahmet Erdem Sariyüce, Kamer Kaya, Erik Saule, and Ümit V. Çatalyürek. Betweenness centrality on GPUs and heterogeneous architectures. In Workshop on General Purpose Processing Using GPUs (GPGPU), in conjunction with ASPLOS, page 10, March 2013. [ bib | slides | .pdf ]
[C45]
Pieter Maris, H Metin Aktulga, Mark A Caprio, Ümit V. Çatalyürek, Edmong Ng, Dossay Oryspayev, Hugh Potter, Erik Saule, Masha Sosonkina, James P Vary, Chao Yang, and Zheng Zhou. Large-scale ab initio configuration interaction calculations for light nuclei. In Journal of Physics: Conference Series. HITES 2012: 'Horizons of Innovative Theories, Experiments, and Supercomputing in Nuclear Physics', volume 403, 2012. [ bib | DOI ]
[C46]
Zheng Zhou, Erik Saule, Hasan Metin Aktulga, Chao Yang, Esmond G. Ng, Pieter Maris, James P. Vary, and Ümit V. Çatalyürek. An out-of-core eigensolver on SSD-equipped clusters. In Proc. of IEEE Cluster, September 2012. [ bib | .pdf ]
[C47]
Onur Küçüktunç, Kamer Kaya, Erik Saule, and Ümit V. Çatalyürek. Fast recommendation on bibliographic networks. In IEEE/ACM International Conference on Social Networks Analysis and Mining (ASONAM), August 2012. acceptance rate: 16%. [ bib | .pdf ]
[C48]
Onur Küçüktunç, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Direction awareness in citation recommendation. In Proceedings of the 6th International Workshop on Ranking in Databases (DBRank), page 6, 2012. [ bib | .pdf ]
[C49]
Zheng Zhou, Erik Saule, Hasan Metin Aktulga, Chao Yang, Esmond G. Ng, Pieter Maris, James P. Vary, and Ümit V. Çatalyürek. An out-of-core dataflow middleware to reduce the cost of large scale iterative solvers. In 2012 International Conference on Parallel Processing (ICPP) Workshops, Fifth International Workshop on Parallel Programming Models and Systems Software for High-End Computing (P2S2), September 2012. [ bib | .pdf ]
[C50]
Kamer Kaya, Erik Saule, Onur Küçüktunç, and Umit Catalyurek. Algorithms for offline tracking of connected components in large evolving networks. In Proceeding of the first SDM Workshop on Dynamic Network Analysis (DNA-SDM), April 2012. [ bib | .pdf ]
[C51]
Ahmet Erdem Sariyüce, Erik Saule, and Umit V. Catalyurek. Scalable hybrid implementation of graph coloring using MPI and OpenMP. In 26th International Symposium on Parallel and Distributed Processing, Workshops and PhD Forum (IPDPSW), Workshop on Parallel Computing and Optimization (PCO), May 2012. [ bib | .pdf ]
[C52]
Erik Saule and Ümit V. Çatalyürek. An early evaluation of the scalability of graph algorithms on the Intel MIC architecture. In 26th International Symposium on Parallel and Distributed Processing, Workshops and PhD Forum (IPDPSW), Workshop on Multithreaded Architectures and Applications (MTAAP), 2012. [ bib | .pdf ]
[C53]
Y.A. Omelchenko, H. Karimabadi, M. Brown, U. V. Catalyurek, and E. Saule. Adaptive multiscale electromagnetic particle simulations. In Bulletin of the American Physical Society, 52nd Annual Meeting of the APS Division of Plasma Physics, Volume 55, Number 15, November 2012. (poster). [ bib ]
[C54]
Ahmet Erdem Sariyüce, Erik Saule, and Umit V. Catalyurek. Improving graph coloring on distributed memory parallel computers. In 18th Annual International Conference on High Performance Computing, 2011. acceptance rate: 19.4%. [ bib | .pdf ]
[C55]
Erik Saule, Erdeniz O. Bas, and Umit V. Catalyurek. Partitioning spatially located computations using rectangles. In the 25th IEEE International Parallel and Distributed Processing Symposium, 2011. acceptance rate: 19.6%. [ bib | slides | .pdf ]
[C56]
Y.A. Omelchenko, H. Karimabadi, M. Brown, U. V. Catalyurek, and E. Saule. Asynchronous multi-dimensional hybrid simulations of magnetized plasmas. In Bulletin of the American Physical Society, 53rd Annual Meeting of the APS Division of Plasma Physics, Volume 56, Number 16, November 2011. (poster). [ bib ]
[C57]
Timothy D. R. Hartley, Erik Saule, and Umit V. Catalyurek. Automatic dataflow application tuning for heterogeneous systems. In Proceedings of The 17th International Conference on High Performance Computing (HiPC 2010), 2010. acceptance rate: 19.2%. [ bib | .pdf ]
[C58]
Florent Blachot, Guillaume Huard, Jonathan Pecero, Erik Saule, and Denis Trystram. Scheduling instructions on hierarchical machines. In The 11th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC 2010), April 2010. [ bib | .pdf ]
[C59]
Erik Saule, Doruk Bozdag, and Ümit V. Çatalyürek. A moldable online scheduling algorithm and its application to parallel short sequence mapping. In Springer, editor, 15th Workshop on Job Scheduling Strategies for Parallel Processing, volume 6253 of LNCS, 2010. [ bib | slides | .pdf ]
[C60]
Y. A. Omelchenko, H. Karimabadi, E. Saule, and U. V. Catalyurek. Parallel Event-Driven Global Magnetospheric Hybrid Simulations. In AGU Fall Meeting Abstracts, page A1756, December 2010. (abstract). [ bib ]
[C61]
Erik Saule and Denis Trystram. Multi-users scheduling in parallel systems. In Proc. of IEEE International Parallel and Distributed Processing Symposium 2009, May 2009. acceptance rate : 22.7%. [ bib | .pdf ]
[C62]
Brice Videau, Erik Saule, and Jean-François Méhaut. Pastel : Parallel runtime and algorithms for small datasets. In proc of MuCoCos, March 2009. [ bib ]
[C63]
Y. Omelchenko, H. Karimabadi, U. Catalyurek, E. Saule, and M. R. Brown. HYPERS: First Ever Multi-Dimensional Asynchronous Hybrid Simulations. page A1316, December 2009. (abstract). [ bib ]
[C64]
Y.A. Omelchenko, H. Karimabadi, M. Brown, U. V. Catalyurek, and E. Saule. Enabling global kinetic simulations of the magnetosphere via petascale computing. In Bulletin of the American Physical Society, 51st Annual Meeting of the APS Division of Plasma Physics, Volume 54, Number 15, November 2009. (poster). [ bib ]
[C65]
Emmanuel Jeannot, Erik Saule, and Denis Trystram. Bi-objective approximation scheme for makespan and reliability optimization on uniform parallel machines. In Euro-Par 2008, volume 5168 of LNCS, pages 877--886. Springer, August 2008. acceptance rate: 33.7%. [ bib | .pdf ]
[C66]
Erik Saule, Pierre-François Dutot, and Gregory Mounié. Scheduling With Storage Constraints. In Proc of IPDPS'08, April 2008. acceptance rate: 25.6%. [ bib | .pdf ]
[C67]
Erik Saule and Brice Videau. PaSTeL. Une implantation parallèle de la STL pour les architectures multi-coeurs : une analyse des performances. In Proceedings électronique de RenPar 18, February 2008. [ bib | .pdf ]
[C68]
Jack J. Dongarra, Emmanuel Jeannot, Erik Saule, and Zhiao Shi. Bi-objective scheduling algorithms for optimizing makespan and reliability on heterogeneous systems. In SPAA '07: Proceedings of the nineteenth annual ACM symposium on Parallelism in algorithms and architectures, pages 280--288. ACM press, June 2007. acceptance rate: 28%. [ bib | .pdf ]
[C69]
Florent Blachot, Guillaume Huard, Jonathan Pecero, Erik Saule, and Denis Trystram. Scheduling instructions on processors with incomplete bypass. In Proceedings of 8th Workshop on Models and Algorithms for Planning and Scheduling Problems. Koç University, July 2007. [ bib | .pdf ]

Tutorials

[Tu1]
Kalpathi Subramanian, Erik Saule, and Jamie Payton. Building engaging assignments for your class. In Proc. of SIGCSE, 2023. SIGCSE workshop. [ bib | DOI ]
[Tu2]
Kalpathi Subramanian, Erik Saule, Matthew Mcquaigue, and Jamie Payton. Improving the structure and content of early cs courses with well aligned, engaging learning materials. In Proc. of SIGCSE, 2022. SIGCSE workshop. [ bib ]
[Tu3]
Erik Saule, Kalpathi Subramanian, and Jamie Payton. Cs materials: A system to assess and align your courses to national standards. In Proc. CCSC:SE, January 2022. [ bib | .pdf ]
[Tu4]
Erik Saule, Alec Goncharow, and Jamie Payton. Using cs materials, a system to align your courses with national standards. In Proceedings of the 52nd ACM Technical Symposium on Computer Science Education, SIGCSE '21, page 1353, New York, NY, USA, 2021. Association for Computing Machinery. SIGCSE workshop. [ bib | DOI ]
[Tu5]
Erik Saule, Kalpathi Subramanian, and Jamie Payton. Using cs materials, a system to align your courses with national standards – conference workshop. In The Journal of ComputingSciences in Colleges, volume 36, October 2020. [ bib ]
[Tu6]
Kalpathi Subramanian, Jamie Payton, and Erik Saule. Bringing real-world data and visualizations of student-implemented data structures into sophomore cs courses using bridges. In Proceedings of the 50th ACM Technical Symposium on Computer Science Education, SIGCSE '19, pages 1235--1235, New York, NY, USA, 2019. ACM. SIGCSE Workshop. [ bib | DOI ]

Technical Reports

[R1]
Haofeng Jia and Erik Saule. Graph embedding for citation recommendation. Technical Report abs/1812.03835, CoRR, 2018. [ bib | arXiv | http ]
[R2]
Erik Saule, Dinesh Panchananam, Alexander Hohl, Wenwu Tang, and Eric Delmelle. Parallel space-time kernel density estimation. Technical Report arXiv:1705.09366, ArXiv, May 2017. [ bib | http | .pdf ]
[R3]
Hugh Potter, Dossay Oryspayev, Pieter Maris, Masha Sosonkina, James Vary, Sven Binder, Angelo Calci, Joachim Langhammer, Robert Roth, Ümit Çatalyürek, and Erik Saule. Accelerating ab initio nuclear physics calculations with gpus. Technical Report arXiv:1412.5989, ArXiv, December 2014. [ bib ]
[R4]
Ahmet Erdem Sariyüce, Erik Saule, and Ümit V. Çatalyürek. On distributed graph coloring with iterative recoloring. Technical Report arXiv:1407.6745, ArXiv, July 2014. [ bib | http ]
[R5]
Ahmet Erdem Sariyüce, Kamer Kaya, Erik Saule, and Ümit V. Çatalyürek. Incremental algorithms for network management and analysis based on closeness centrality. Technical Report arXiv:1303.0422, ArXiv, February 2013. [ bib | http | .pdf ]
[R6]
Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Performance evaluation of sparse matrix multiplication kernels on intel xeon phi. Technical Report arXiv:1302.1078, ArXiv, February 2013. [ bib | http | .pdf ]
[R7]
Ahmet Erdem Sariyüce, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Shattering and compressing networks for centrality analysis. Technical Report arXiv:1209.6007, ArXiv, September 2012. [ bib | http | .pdf ]
[R8]
Onur Küçüktunç, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Diversifying citation recommendations. Technical Report arXiv:1209.5809, ArXiv, September 2012. [ bib | http | .pdf ]
[R9]
Onur Küçüktunç, Erik Saule, Kamer Kaya, and Ümit V. Çatalyürek. Recommendation on academic networks using direction aware citation analysis. Technical Report arXiv:1205.1143, ArXiv, April 2012. [ bib | http | .pdf ]
[R10]
Erik Saule, Erdeniz O. Bas, and Ümit V. Çatalyürek. Load-balancing spatially located computations using rectangular partitions. Technical Report arXiv:1104.2566v1, ArXiv, April 2011. [ bib | http | .pdf ]
[R11]
Anne Benoit, Umit Catalyurek, Yves Robert, and Erik Saule. A Survey of Pipelined Workflow Scheduling: Models and Algorithms. Technical Report RR-LIP-2010-28, LIP, ENS Lyon, France, September 2010. [ bib | http | .pdf ]
[R12]
Brice Videau, Erik Saule, and Jean-François Méhaut. PaSTeL : Parallel Runtime and Algorithms for Small Datasets. Technical Report 6650, INRIA, 2008. [ bib | http ]

Under Review

Invited Talks

[T1]
Erik Saule. Optimizing distributed graph coloring. Inria Bordeaux Seminar, May 2024. [ bib | slides ]
[T2]
Erik Saule. Optimizing distributed graph coloring. New Challenges in Scheduling Theory, May 2024. [ bib | slides ]
[T3]
Erik Saule. Coloring graphs with intervals. The 16th Scheduling for Large Scale Systems Workshop, May 2023. [ bib | slides ]
[T4]
Erik Saule. Parallel space-time kernel density estimation. LIG Seminar, April 2018. [ bib | slides ]
[T5]
Erik Saule. Parallel space-time kernel density estimation. The 12th Scheduling for Large Scale Systems Workshop, May 2017. [ bib | slides ]
[T6]
Erik Saule. Computing graph centrality. SIAM CSE 2017, March 2017. [ bib | slides ]
[T7]
Erik Saule. Time-cost trade-offs of pipelined dataflow applications. New Challenges in Scheduling Theory, April 2016. [ bib | slides ]
[T8]
Erik Saule. Centrality of evolving networks. IEEE seminar at North Carolina A&T State University, November 2015. [ bib | slides ]
[T9]
Erik Saule. GPU-accelerated network centrality. GPU Technology Conference, March 2015. [ bib | slides ]
[T10]
Erik Saule. Parallel dataflow graph coloring. Dagstuhl seminar on Algorithms and Scheduling Techniques for Exascale Systems, September 2013. [ bib | slides ]
[T11]
Erik Saule. Large scale graph analysis. UMass Boston seminar, March 2013. [ bib | slides ]
[T12]
Erik Saule. Large scale graph analysis. UNCC seminar, February 2013. [ bib | slides ]
[T13]
Erik Saule. Academic recommendation using citation analysis with theAdvisor. Keynote talk at "Computational Scientometics: Theory and Applications", February 2013. [ bib | slides ]
[T14]
Erik Saule. Partitioning spatially located load with rectangles: Algorithms and simulations. New Challenges on Scheduling Theory, September 2010. [ bib | slides ]
[T15]
Erik Saule. Optimizing the maximum stretch of online tasks on a parallel system without preemption. 3rd ”Scheduling in Aussois” Workshop, June 2010. [ bib | slides ]
[T16]
Erik Saule. Load balancing of spatially located computation - the one dimensional case. GRAAL group meeting, March 2010. [ bib ]
[T17]
Erik Saule. Multi-objective optimization/approximation in scheduling. Workshop/Summer school on Algorithms and Techniques for Scheduling on Clusters and Grids (ASTEC), June 2009. [ bib ]
[T18]
Erik Saule. A moldable online scheduling algorithm and its application to parallel short sequence mapping. Scheduling for large-scale systems, May 2009. [ bib ]
[T19]
Erik Saule. The multi user scheduling problem. Journée GOThA, January 2009. [ bib ]
[T20]
Erik Saule. User centered scheduling for multi-users parallel systems. New Challenges on Scheduling Theory, May 2008. [ bib ]
[T21]
Erik Saule. Scheduling instructions on processors with incomplete bypass. Journée GOThA, April 2008. [ bib ]
[T22]
Erik Saule. Scheduling instructions on processors with incomplete bypass. IPIPAN hpc group meeting, December 2007. [ bib ]
[T23]
Emmanuel Jeannot and Erik Saule. Ordonnancement sur machine heterogene a but de makespan et de fiabilite. Journée Mao, January 2007. [ bib ]
[T24]
Erik Saule. Reliability versus performance for embedded real-time applications. Scheduling Algorithms for new Emerging Applications, June 2006. [ bib ]
[T25]
Erik Saule. Une approche d'équilibrage de charge par théorie des jeux. Journées Théorie des jeux AlGorithmique et Applications Dans les réseAux (TAGADA), February 2006. [ bib ]