Christian Godenschwager, F. Schornbaum, Martin Bauer, H. Köstler, U. Rüde
waLBerla is a massively parallel software framework for simulating complex flows with the lattice Boltzmann method (LBM). Performance and scalability results are presented for SuperMUC, the world's fastest x86-based supercomputer ranked number 6 on the Top500 list, and JUQUEEN, a Blue Gene/Q system ranked as number 5. We reach resolutions with more than one trillion cells and perform up to 1.93 trillion cell updates per second using 1.8 million threads. The design and implementation of waLBerla is driven by a careful analysis of the performance on current petascale supercomputers. Our fully distributed data structures and algorithms allow for efficient, massively parallel simulations on these machines. Elaborate node level optimizations and vectorization using SIMD instructions result in highly optimized compute kernels for the single- and two-relaxation-time LBM. Excellent weak and strong scaling is achieved for a complex vascular geometry of the human coronary tree.
{"title":"A framework for hybrid parallel flow simulations with a trillion cells in complex geometries","authors":"Christian Godenschwager, F. Schornbaum, Martin Bauer, H. Köstler, U. Rüde","doi":"10.1145/2503210.2503273","DOIUrl":"https://doi.org/10.1145/2503210.2503273","url":null,"abstract":"waLBerla is a massively parallel software framework for simulating complex flows with the lattice Boltzmann method (LBM). Performance and scalability results are presented for SuperMUC, the world's fastest x86-based supercomputer ranked number 6 on the Top500 list, and JUQUEEN, a Blue Gene/Q system ranked as number 5. We reach resolutions with more than one trillion cells and perform up to 1.93 trillion cell updates per second using 1.8 million threads. The design and implementation of waLBerla is driven by a careful analysis of the performance on current petascale supercomputers. Our fully distributed data structures and algorithms allow for efficient, massively parallel simulations on these machines. Elaborate node level optimizations and vectorization using SIMD instructions result in highly optimized compute kernels for the single- and two-relaxation-time LBM. Excellent weak and strong scaling is achieved for a complex vascular geometry of the human coronary tree.","PeriodicalId":371074,"journal":{"name":"2013 SC - International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130833170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Tong Jin, Fan Zhang, Qian Sun, H. Bui, M. Parashar, Hongfeng Yu, S. Klasky, N. Podhorszki, H. Abbasi
As system scales and application complexity grow, managing and processing simulation data has become a significant challenge. While recent approaches based on data staging and in-situ/in-transit data processing are promising, dynamic data volumes and distributions,such as those occurring in AMR-based simulations, make the efficient use of these techniques challenging. In this paper we propose cross-layer adaptations that address these challenges and respond at runtime to dynamic data management requirements. Specifically we explore (1) adaptations of the spatial resolution at which the data is processed, (2) dynamic placement and scheduling of data processing kernels, and (3) dynamic allocation of in-transit resources. We also exploit co-ordinated approaches that dynamically combine these adaptations at the different layers. We evaluate the performance of our adaptive cross-layer management approach on the Intrepid IBM-BlueGene/P and Titan Cray-XK7 systems using Chombo-based AMR applications, and demonstrate its effectiveness in improving overall time-to-solution and increasing resource efficiency.
{"title":"Using cross-layer adaptations for dynamic data management in large scale coupled scientific workflows","authors":"Tong Jin, Fan Zhang, Qian Sun, H. Bui, M. Parashar, Hongfeng Yu, S. Klasky, N. Podhorszki, H. Abbasi","doi":"10.1145/2503210.2503301","DOIUrl":"https://doi.org/10.1145/2503210.2503301","url":null,"abstract":"As system scales and application complexity grow, managing and processing simulation data has become a significant challenge. While recent approaches based on data staging and in-situ/in-transit data processing are promising, dynamic data volumes and distributions,such as those occurring in AMR-based simulations, make the efficient use of these techniques challenging. In this paper we propose cross-layer adaptations that address these challenges and respond at runtime to dynamic data management requirements. Specifically we explore (1) adaptations of the spatial resolution at which the data is processed, (2) dynamic placement and scheduling of data processing kernels, and (3) dynamic allocation of in-transit resources. We also exploit co-ordinated approaches that dynamically combine these adaptations at the different layers. We evaluate the performance of our adaptive cross-layer management approach on the Intrepid IBM-BlueGene/P and Titan Cray-XK7 systems using Chombo-based AMR applications, and demonstrate its effectiveness in improving overall time-to-solution and increasing resource efficiency.","PeriodicalId":371074,"journal":{"name":"2013 SC - International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125809862","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yufei Ren, Tan Li, Dantong Yu, Shudong Jin, T. Robertazzi
Data-intensive applications place stringent requirements on the performance of both back-end storage systems and frontend network interfaces. However, for ultra high-speed data transfer, for example, at 100 Gbps and higher, the effects of multiple bottlenecks along a full end-to-end path, have not been resolved efficiently. In this paper, we describe our implementation of an end-to-end data transfer software at such high-speeds. At the back-end, we construct a storage area network with the iSCSI protocols, and utilize efficient RDMA technology. At the front-end, we design network communication software to transfer data in parallel, and utilize NUMA techniques to maximize the performance of multiple network interfaces. We demonstrate that our system can deliver the full 100 Gbps end-to-end data transfer throughput. The software product is tested rigorously and demonstrated applicable to supporting various data-intensive applications that constantly move bulk data within and across data centers.
{"title":"Design and performance evaluation of NUMA-aware RDMA-based end-to-end data transfer systems","authors":"Yufei Ren, Tan Li, Dantong Yu, Shudong Jin, T. Robertazzi","doi":"10.1145/2503210.2503260","DOIUrl":"https://doi.org/10.1145/2503210.2503260","url":null,"abstract":"Data-intensive applications place stringent requirements on the performance of both back-end storage systems and frontend network interfaces. However, for ultra high-speed data transfer, for example, at 100 Gbps and higher, the effects of multiple bottlenecks along a full end-to-end path, have not been resolved efficiently. In this paper, we describe our implementation of an end-to-end data transfer software at such high-speeds. At the back-end, we construct a storage area network with the iSCSI protocols, and utilize efficient RDMA technology. At the front-end, we design network communication software to transfer data in parallel, and utilize NUMA techniques to maximize the performance of multiple network interfaces. We demonstrate that our system can deliver the full 100 Gbps end-to-end data transfer throughput. The software product is tested rigorously and demonstrated applicable to supporting various data-intensive applications that constantly move bulk data within and across data centers.","PeriodicalId":371074,"journal":{"name":"2013 SC - International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130622987","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Basic Liner algebra subprograms (BLAS) is a fundamental library in scientific computing. In this paper, we present a template-based optimization framework, AUGEM, which can automatically generate fully optimized assembly code for several dense linear algebra (DLA) kernels, such as GEMM, GEMV, AXPY and DOT, on varying multi-core CPUs without requiring any manual interference from developers. In particular, based on domain-specific knowledge about algorithms of the DLA kernels, we use a collection of parameterized code templates to formulate a number of commonly occurring instruction sequences within the optimized low-level C code of these DLA kernels. Then, our framework uses a specialized low-level C optimizer to identify instruction sequences that match the pre-defined code templates and thereby translates them into extremely efficient SSE/AVX instructions. The DLA kernels generated by our templatebased approach surpass the implementations of Intel MKL and AMD ACML BLAS libraries, on both Intel Sandy Bridge and AMD Piledriver processors.
{"title":"AUGEM: Automatically generate high performance Dense Linear Algebra kernels on x86 CPUs","authors":"Qian Wang, Xianyi Zhang, Yunquan Zhang, Qing Yi","doi":"10.1145/2503210.2503219","DOIUrl":"https://doi.org/10.1145/2503210.2503219","url":null,"abstract":"Basic Liner algebra subprograms (BLAS) is a fundamental library in scientific computing. In this paper, we present a template-based optimization framework, AUGEM, which can automatically generate fully optimized assembly code for several dense linear algebra (DLA) kernels, such as GEMM, GEMV, AXPY and DOT, on varying multi-core CPUs without requiring any manual interference from developers. In particular, based on domain-specific knowledge about algorithms of the DLA kernels, we use a collection of parameterized code templates to formulate a number of commonly occurring instruction sequences within the optimized low-level C code of these DLA kernels. Then, our framework uses a specialized low-level C optimizer to identify instruction sequences that match the pre-defined code templates and thereby translates them into extremely efficient SSE/AVX instructions. The DLA kernels generated by our templatebased approach surpass the implementations of Intel MKL and AMD ACML BLAS libraries, on both Intel Sandy Bridge and AMD Piledriver processors.","PeriodicalId":371074,"journal":{"name":"2013 SC - International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130693930","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
George Michelogiannakis, Nan Jiang, Daniel U. Becker, W. Dally
Channels in system-wide networks tend to be over-subscribed due to the cost of bandwidth and increasing traffic demands. To make matters worse, workloads can overstress specific destinations, creating hotspots. Lossless networks offer attractive advantages compared to lossy networks but suffer from tree saturation. This led to the development of explicit congestion notification (ECN). However, ECN is very sensitive to its configuration parameters and acts only after congestion forms. We propose channel reservation protocol (CRP) to enable sources to reserve bandwidth in multiple resources in advance of packet transmission and with a single request, but without idling resources like circuit switching. CRP prevents congestion from ever occurring and thus reacts instantly to traffic changes, whereas ECN requires 300,000 cycles to stabilize in our experiments. Furthermore, ECN may not prevent congestion formed by short-lived flows generated by a large combination of source-destination pairs.
{"title":"Channel reservation protocol for over-subscribed channels and destinations","authors":"George Michelogiannakis, Nan Jiang, Daniel U. Becker, W. Dally","doi":"10.1145/2503210.2503213","DOIUrl":"https://doi.org/10.1145/2503210.2503213","url":null,"abstract":"Channels in system-wide networks tend to be over-subscribed due to the cost of bandwidth and increasing traffic demands. To make matters worse, workloads can overstress specific destinations, creating hotspots. Lossless networks offer attractive advantages compared to lossy networks but suffer from tree saturation. This led to the development of explicit congestion notification (ECN). However, ECN is very sensitive to its configuration parameters and acts only after congestion forms. We propose channel reservation protocol (CRP) to enable sources to reserve bandwidth in multiple resources in advance of packet transmission and with a single request, but without idling resources like circuit switching. CRP prevents congestion from ever occurring and thus reacts instantly to traffic changes, whereas ECN requires 300,000 cycles to stabilize in our experiments. Furthermore, ECN may not prevent congestion formed by short-lived flows generated by a large combination of source-destination pairs.","PeriodicalId":371074,"journal":{"name":"2013 SC - International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129684254","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Md. Mostofa Ali Patwary, Diana Palsetia, Ankit Agrawal, W. Liao, F. Manne, A. Choudhary
OPTICS is a hierarchical density-based data clustering algorithm that discovers arbitrary-shaped clusters and eliminates noise using adjustable reachability distance thresholds. Parallelizing OPTICS is considered challenging as the algorithm exhibits a strongly sequential data access order. We present a scalable parallel OPTICS algorithm (POPTICS) designed using graph algorithmic concepts. To break the data access sequentiality, POPTICS exploits the similarities between the OPTICS algorithm and PRIM's Minimum Spanning Tree algorithm. Additionally, we use the disjoint-set data structure to achieve a high parallelism for distributed cluster extraction. Using high dimensional datasets containing up to a billion floating point numbers, we show scalable speedups of up to 27.5 for our OpenMP implementation on a 40-core shared-memory machine, and up to 3,008 for our MPI implementation on a 4,096-core distributed-memory machine. We also show that the quality of the results given by POPTICS is comparable to those given by the classical OPTICS algorithm.
{"title":"Scalable parallel OPTICS data clustering using graph algorithmic techniques","authors":"Md. Mostofa Ali Patwary, Diana Palsetia, Ankit Agrawal, W. Liao, F. Manne, A. Choudhary","doi":"10.1145/2503210.2503255","DOIUrl":"https://doi.org/10.1145/2503210.2503255","url":null,"abstract":"OPTICS is a hierarchical density-based data clustering algorithm that discovers arbitrary-shaped clusters and eliminates noise using adjustable reachability distance thresholds. Parallelizing OPTICS is considered challenging as the algorithm exhibits a strongly sequential data access order. We present a scalable parallel OPTICS algorithm (POPTICS) designed using graph algorithmic concepts. To break the data access sequentiality, POPTICS exploits the similarities between the OPTICS algorithm and PRIM's Minimum Spanning Tree algorithm. Additionally, we use the disjoint-set data structure to achieve a high parallelism for distributed cluster extraction. Using high dimensional datasets containing up to a billion floating point numbers, we show scalable speedups of up to 27.5 for our OpenMP implementation on a 40-core shared-memory machine, and up to 3,008 for our MPI implementation on a 4,096-core distributed-memory machine. We also show that the quality of the results given by POPTICS is comparable to those given by the classical OPTICS algorithm.","PeriodicalId":371074,"journal":{"name":"2013 SC - International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"220 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122524265","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
S. Saini, Haoqiang Jin, D. Jespersen, Huiyu Feng, M. J. Djomehri, William Arasin, R. Hood, P. Mehrotra, R. Biswas
Intel recently introduced the Xeon Phi coprocessor based on the Many Integrated Core architecture featuring 60 cores with a peak performance of 1.0 Tflop/s. NASA has deployed a 128-node SGI Rackable system where each node has two Intel Xeon E2670 8-core Sandy Bridge processors along with two Xeon Phi 5110P coprocessors. We have conducted an early performance evaluation of the Xeon Phi. We used microbenchmarks to measure the latency and bandwidth of memory and interconnect, I/O rates, and the performance of OpenMP directives and MPI functions. We also used OpenMP and MPI versions of the NAS Parallel Benchmarks along with two production CFD applications to test four programming modes: offload, processor native, coprocessor native and symmetric (processor plus coprocessor). In this paper we present preliminary results based on our performance evaluation of various aspects of a Phi-based system.
{"title":"An early performance evaluation of many integrated core architecture based sgi rackable computing system","authors":"S. Saini, Haoqiang Jin, D. Jespersen, Huiyu Feng, M. J. Djomehri, William Arasin, R. Hood, P. Mehrotra, R. Biswas","doi":"10.1145/2503210.2503272","DOIUrl":"https://doi.org/10.1145/2503210.2503272","url":null,"abstract":"Intel recently introduced the Xeon Phi coprocessor based on the Many Integrated Core architecture featuring 60 cores with a peak performance of 1.0 Tflop/s. NASA has deployed a 128-node SGI Rackable system where each node has two Intel Xeon E2670 8-core Sandy Bridge processors along with two Xeon Phi 5110P coprocessors. We have conducted an early performance evaluation of the Xeon Phi. We used microbenchmarks to measure the latency and bandwidth of memory and interconnect, I/O rates, and the performance of OpenMP directives and MPI functions. We also used OpenMP and MPI versions of the NAS Parallel Benchmarks along with two production CFD applications to test four programming modes: offload, processor native, coprocessor native and symmetric (processor plus coprocessor). In this paper we present preliminary results based on our performance evaluation of various aspects of a Phi-based system.","PeriodicalId":371074,"journal":{"name":"2013 SC - International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126262923","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Due to the enormous energy consumption and associated environmental concerns, data centers have been increasingly pressured to reduce long-term net carbon footprint to zero, i.e., carbon neutrality. In this paper, we propose an online algorithm, called COCA (optimizing for COst minimization and CArbon neutrality), for minimizing data center operational cost while satisfying carbon neutrality without long-term future information. Unlike the existing research, COCA enables distributed server-level resource management: each server autonomously adjusts its processing speed and optimally decides the amount of workloads to process. We prove that COCA achieves a close-to-minimum operational cost (incorporating both electricity and delay costs) compared to the optimal algorithm with future information, while bounding the potential violation of carbon neutrality. We also perform trace-based simulation studies to complement the analysis, and the results show that COCA reduces cost by more than 25% (compared to state of the art) while resulting in a smaller carbon footprint.
{"title":"COCA: Online distributed resource management for cost minimization and carbon neutrality in data centers","authors":"Shaolei Ren, Yuxiong He","doi":"10.1145/2503210.2503248","DOIUrl":"https://doi.org/10.1145/2503210.2503248","url":null,"abstract":"Due to the enormous energy consumption and associated environmental concerns, data centers have been increasingly pressured to reduce long-term net carbon footprint to zero, i.e., carbon neutrality. In this paper, we propose an online algorithm, called COCA (optimizing for COst minimization and CArbon neutrality), for minimizing data center operational cost while satisfying carbon neutrality without long-term future information. Unlike the existing research, COCA enables distributed server-level resource management: each server autonomously adjusts its processing speed and optimally decides the amount of workloads to process. We prove that COCA achieves a close-to-minimum operational cost (incorporating both electricity and delay costs) compared to the optimal algorithm with future information, while bounding the potential violation of carbon neutrality. We also perform trace-based simulation studies to complement the analysis, and the results show that COCA reduces cost by more than 25% (compared to state of the art) while resulting in a smaller carbon footprint.","PeriodicalId":371074,"journal":{"name":"2013 SC - International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"85 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129318365","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Near the dawn of the petascale era, IO libraries had reached a stability in their function and data layout with only incremental changes being incorporated. The shift in technology, particularly the scale of parallel file systems and the number of compute processes, prompted revisiting best practices for optimal IO performance. Among other efforts like PLFS, the project that led to ADIOS, the ADaptable IO System, was motivated by both the shift in technology and the historical requirement, for optimal IO performance, to change how simulations performed IO depending on the platform. To solve both issues, the ADIOS team, along with consultation with other leading IO experts, sought to build a new IO platform based on the assumptions inherent in the petascale hardware platforms. This paper helps inform the design of future IO platforms with a discussion of lessons learned as part of the process of designing and building ADIOS.
{"title":"Insights for exascale IO APIs from building a petascale IO API","authors":"J. Lofstead, R. Ross","doi":"10.1145/2503210.2503238","DOIUrl":"https://doi.org/10.1145/2503210.2503238","url":null,"abstract":"Near the dawn of the petascale era, IO libraries had reached a stability in their function and data layout with only incremental changes being incorporated. The shift in technology, particularly the scale of parallel file systems and the number of compute processes, prompted revisiting best practices for optimal IO performance. Among other efforts like PLFS, the project that led to ADIOS, the ADaptable IO System, was motivated by both the shift in technology and the historical requirement, for optimal IO performance, to change how simulations performed IO depending on the platform. To solve both issues, the ADIOS team, along with consultation with other leading IO experts, sought to build a new IO platform based on the assumptions inherent in the petascale hardware platforms. This paper helps inform the design of future IO platforms with a discussion of lessons learned as part of the process of designing and building ADIOS.","PeriodicalId":371074,"journal":{"name":"2013 SC - International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"26 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132273519","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pai-Wei Lai, Kevin Stock, Samyam Rajbhandari, S. Krishnamoorthy, P. Sadayappan
In this paper, we introduce the Dynamic Load-balanced Tensor Contractions (DLTC), a domain-specific library for efficient task parallel execution of tensor contraction expressions, a class of computation encountered in quantum chemistry and physics. Our framework decomposes each contraction into smaller unit of tasks, represented by an abstraction referred to as iterators. We exploit an extra level of parallelism by having tasks across independent contractions executed concurrently through a dynamic load balancing runtime. We demonstrate the improved performance, scalability, and flexibility for the computation of tensor contraction expressions on parallel computers using examples from Coupled Cluster (CC) methods.
{"title":"A framework for load balancing of Tensor Contraction expressions via dynamic task partitioning","authors":"Pai-Wei Lai, Kevin Stock, Samyam Rajbhandari, S. Krishnamoorthy, P. Sadayappan","doi":"10.1145/2503210.2503290","DOIUrl":"https://doi.org/10.1145/2503210.2503290","url":null,"abstract":"In this paper, we introduce the Dynamic Load-balanced Tensor Contractions (DLTC), a domain-specific library for efficient task parallel execution of tensor contraction expressions, a class of computation encountered in quantum chemistry and physics. Our framework decomposes each contraction into smaller unit of tasks, represented by an abstraction referred to as iterators. We exploit an extra level of parallelism by having tasks across independent contractions executed concurrently through a dynamic load balancing runtime. We demonstrate the improved performance, scalability, and flexibility for the computation of tensor contraction expressions on parallel computers using examples from Coupled Cluster (CC) methods.","PeriodicalId":371074,"journal":{"name":"2013 SC - International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131376762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：