J. Chhugani, Changkyu Kim, H. Shukla, Jongsoo Park, P. Dubey, J. Shalf, H. Simon
Two-point Correlation Function (TPCF) is widely used in astronomy to characterize the distribution of matter/energy in the Universe, and help derive the physics that can trace back to the creation of the universe. However, it is prohibitively slow for current sized datasets, and would continue to be a critical bottleneck with the trend of increasing dataset sizes to billions of particles and more, which makes TPCF a compelling benchmark application for future exa-scale architectures. State-of-the-art TPCF implementations do not map well to the underlying SIMD hardware, and also suffer from load-imbalance for large core counts. In this paper, we present a novel SIMD-friendly histogram update algorithm that exploits the spatial locality of histogram updates to achieve near-linear SIMD scaling. We also present a load-balancing scheme that combines domain-specific initial static division of work and dynamic task migration across nodes to effectively balance computation across nodes. Using Zin supercomputer at Lawrence Livermore National Laboratory (25,600 cores of Intel® Xeon® E5-2670, each with 256-bit SIMD), we achieve 90% parallel efficiency and 96% SIMD efficiency, and perform TPCF computation on a 1.7 billion particle dataset in 5.3 hours (at least 35× faster than previous approaches). In terms of cost per performance (measured in flops/$), we achieve at least an order-of-magnitude (11.1x) higher flops/$ as compared to the best known results [1]. Consequently, we now have line-of-sight to achieving the processing power for correlation computation to process billion+ particles telescopic data.
{"title":"Billion-particle SIMD-friendly two-point correlation on large-scale HPC cluster systems","authors":"J. Chhugani, Changkyu Kim, H. Shukla, Jongsoo Park, P. Dubey, J. Shalf, H. Simon","doi":"10.1109/SC.2012.24","DOIUrl":"https://doi.org/10.1109/SC.2012.24","url":null,"abstract":"Two-point Correlation Function (TPCF) is widely used in astronomy to characterize the distribution of matter/energy in the Universe, and help derive the physics that can trace back to the creation of the universe. However, it is prohibitively slow for current sized datasets, and would continue to be a critical bottleneck with the trend of increasing dataset sizes to billions of particles and more, which makes TPCF a compelling benchmark application for future exa-scale architectures. State-of-the-art TPCF implementations do not map well to the underlying SIMD hardware, and also suffer from load-imbalance for large core counts. In this paper, we present a novel SIMD-friendly histogram update algorithm that exploits the spatial locality of histogram updates to achieve near-linear SIMD scaling. We also present a load-balancing scheme that combines domain-specific initial static division of work and dynamic task migration across nodes to effectively balance computation across nodes. Using Zin supercomputer at Lawrence Livermore National Laboratory (25,600 cores of Intel® Xeon® E5-2670, each with 256-bit SIMD), we achieve 90% parallel efficiency and 96% SIMD efficiency, and perform TPCF computation on a 1.7 billion particle dataset in 5.3 hours (at least 35× faster than previous approaches). In terms of cost per performance (measured in flops/$), we achieve at least an order-of-magnitude (11.1x) higher flops/$ as compared to the best known results [1]. Consequently, we now have line-of-sight to achieving the processing power for correlation computation to process billion+ particles telescopic data.","PeriodicalId":217221,"journal":{"name":"2012 International Conference for High Performance Computing, Networking, Storage and Analysis","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134130637","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}
B. Nouanesengsy, Teng-Yok Lee, Kewei Lu, Han-Wei Shen, T. Peterka
Flow fields are an important product of scientific simulations. One popular flow visualization technique is particle advection, in which seeds are traced through the flow field. One use of these traces is to compute a powerful analysis tool called the Finite-Time Lyapunov Exponent (FTLE) field, but no existing particle tracing algorithms scale to the particle injection frequency required for high-resolution FTLE analysis. In this paper, a framework to trace the massive number of particles necessary for FTLE computation is presented. A new approach is explored, in which processes are divided into groups, and are responsible for mutually exclusive spans of time. This pipelining over time intervals reduces overall idle time of processes and decreases I/O overhead. Our parallel FTLE framework is capable of advecting hundreds of millions of particles at once, with performance scaling up to tens of thousands of processes.
{"title":"Parallel particle advection and FTLE computation for time-varying flow fields","authors":"B. Nouanesengsy, Teng-Yok Lee, Kewei Lu, Han-Wei Shen, T. Peterka","doi":"10.1109/SC.2012.93","DOIUrl":"https://doi.org/10.1109/SC.2012.93","url":null,"abstract":"Flow fields are an important product of scientific simulations. One popular flow visualization technique is particle advection, in which seeds are traced through the flow field. One use of these traces is to compute a powerful analysis tool called the Finite-Time Lyapunov Exponent (FTLE) field, but no existing particle tracing algorithms scale to the particle injection frequency required for high-resolution FTLE analysis. In this paper, a framework to trace the massive number of particles necessary for FTLE computation is presented. A new approach is explored, in which processes are divided into groups, and are responsible for mutually exclusive spans of time. This pipelining over time intervals reduces overall idle time of processes and decreases I/O overhead. Our parallel FTLE framework is capable of advecting hundreds of millions of particles at once, with performance scaling up to tens of thousands of processes.","PeriodicalId":217221,"journal":{"name":"2012 International Conference for High Performance Computing, Networking, Storage and Analysis","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131851156","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}
Nusrat S. Islam, Md. Wasi-ur-Rahman, Jithin Jose, R. Rajachandrasekar, Hao Wang, H. Subramoni, Chet Murthy, D. Panda
Hadoop Distributed File System (HDFS) acts as the primary storage of Hadoop and has been adopted by reputed organizations (Facebook, Yahoo! etc.) due to its portability and fault-tolerance. The existing implementation of HDFS uses Javasocket interface for communication which delivers suboptimal performance in terms of latency and throughput. For dataintensive applications, network performance becomes key component as the amount of data being stored and replicated to HDFS increases. In this paper, we present a novel design of HDFS using Remote Direct Memory Access (RDMA) over InfiniBand via JNI interfaces. Experimental results show that, for 5GB HDFS file writes, the new design reduces the communication time by 87% and 30% over 1Gigabit Ethernet (1GigE) and IP-over-InfiniBand (IPoIB), respectively, on QDR platform (32Gbps). For HBase, the Put operation performance is improved by 26% with our design. To the best of our knowledge, this is the first design of HDFS over InfiniBand networks.
{"title":"High performance RDMA-based design of HDFS over InfiniBand","authors":"Nusrat S. Islam, Md. Wasi-ur-Rahman, Jithin Jose, R. Rajachandrasekar, Hao Wang, H. Subramoni, Chet Murthy, D. Panda","doi":"10.1109/SC.2012.65","DOIUrl":"https://doi.org/10.1109/SC.2012.65","url":null,"abstract":"Hadoop Distributed File System (HDFS) acts as the primary storage of Hadoop and has been adopted by reputed organizations (Facebook, Yahoo! etc.) due to its portability and fault-tolerance. The existing implementation of HDFS uses Javasocket interface for communication which delivers suboptimal performance in terms of latency and throughput. For dataintensive applications, network performance becomes key component as the amount of data being stored and replicated to HDFS increases. In this paper, we present a novel design of HDFS using Remote Direct Memory Access (RDMA) over InfiniBand via JNI interfaces. Experimental results show that, for 5GB HDFS file writes, the new design reduces the communication time by 87% and 30% over 1Gigabit Ethernet (1GigE) and IP-over-InfiniBand (IPoIB), respectively, on QDR platform (32Gbps). For HBase, the Put operation performance is improved by 26% with our design. To the best of our knowledge, this is the first design of HDFS over InfiniBand networks.","PeriodicalId":217221,"journal":{"name":"2012 International Conference for High Performance Computing, Networking, Storage and Analysis","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132647108","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}
An important challenge faced by high-throughput, multiscale applications is that human intervention has a central role in driving their success. However, manual intervention is inefficient, error-prone and promotes resource wasting. This paper presents an application-aware modular framework that provides self-management for computational multiscale applications in volunteer computing (VC). Our framework consists of a learning engine and three modules that can be easily adapted to different distributed systems. The learning engine of this framework is based on our novel tree-like structure called KOTree. KOTree is a fully automatic method that organizes statistical information in a multidimensional structure that can be efficiently searched and updated at runtime. Our empirical evaluation shows that our framework can effectively provide application-aware self-management in VC systems. Additionally, we observed that our KOTree algorithm is able to predict accurately the expected length of new jobs, resulting in an average of 85% increased throughput with respect to other algorithms.
{"title":"On the effectiveness of application-aware self-management for scientific discovery in volunteer computing systems","authors":"Trilce Estrada, M. Taufer","doi":"10.1109/SC.2012.83","DOIUrl":"https://doi.org/10.1109/SC.2012.83","url":null,"abstract":"An important challenge faced by high-throughput, multiscale applications is that human intervention has a central role in driving their success. However, manual intervention is inefficient, error-prone and promotes resource wasting. This paper presents an application-aware modular framework that provides self-management for computational multiscale applications in volunteer computing (VC). Our framework consists of a learning engine and three modules that can be easily adapted to different distributed systems. The learning engine of this framework is based on our novel tree-like structure called KOTree. KOTree is a fully automatic method that organizes statistical information in a multidimensional structure that can be efficiently searched and updated at runtime. Our empirical evaluation shows that our framework can effectively provide application-aware self-management in VC systems. Additionally, we observed that our KOTree algorithm is able to predict accurately the expected length of new jobs, resulting in an average of 85% increased throughput with respect to other algorithms.","PeriodicalId":217221,"journal":{"name":"2012 International Conference for High Performance Computing, Networking, Storage and Analysis","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124717122","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}
Prediction of host load in Cloud systems is critical for achieving service-level agreements. However, accurate prediction of host load in Clouds is extremely challenging because it fluctuates drastically at small timescales. We design a prediction method based on Bayes model to predict the mean load over a long-term time interval, as well as the mean load in consecutive future time intervals. We identify novel predictive features of host load that capture the expectation, predictability, trends and patterns of host load. We also determine the most effective combinations of these features for prediction. We evaluate our method using a detailed one-month trace of a Google data center with thousands of machines. Experiments show that the Bayes method achieves high accuracy with a mean squared error of 0.0014. Moreover, the Bayes method improves the load prediction accuracy by 5.6 -- 50% compared to other state-of-the-art methods based on moving averages, auto-regression, and/or noise filters.
{"title":"Host load prediction in a Google compute cloud with a Bayesian model","authors":"S. Di, Derrick Kondo, W. Cirne","doi":"10.1109/SC.2012.68","DOIUrl":"https://doi.org/10.1109/SC.2012.68","url":null,"abstract":"Prediction of host load in Cloud systems is critical for achieving service-level agreements. However, accurate prediction of host load in Clouds is extremely challenging because it fluctuates drastically at small timescales. We design a prediction method based on Bayes model to predict the mean load over a long-term time interval, as well as the mean load in consecutive future time intervals. We identify novel predictive features of host load that capture the expectation, predictability, trends and patterns of host load. We also determine the most effective combinations of these features for prediction. We evaluate our method using a detailed one-month trace of a Google data center with thousands of machines. Experiments show that the Bayes method achieves high accuracy with a mean squared error of 0.0014. Moreover, the Bayes method improves the load prediction accuracy by 5.6 -- 50% compared to other state-of-the-art methods based on moving averages, auto-regression, and/or noise filters.","PeriodicalId":217221,"journal":{"name":"2012 International Conference for High Performance Computing, Networking, Storage and Analysis","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130543226","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}
Preeti Malakar, Thomas George, Sameer Kumar, R. Mittal, V. Natarajan, Yogish Sabharwal, Vaibhav Saxena, Sathish S. Vadhiyar
Accurate and timely prediction of weather phenomena, such as hurricanes and flash floods, require high-fidelity compute intensive simulations of multiple finer regions of interest within a coarse simulation domain. Current weather applications execute these nested simulations sequentially using all the available processors, which is sub-optimal due to their sub-linear scalability. In this work, we present a strategy for parallel execution of multiple nested domain simulations based on partitioning the 2-D processor grid into disjoint rectangular regions associated with each domain. We propose a novel combination of performance prediction, processor allocation methods and topology-aware mapping of the regions on torus interconnects. Experiments on IBM Blue Gene systems using WRF show that the proposed strategies result in performance improvement of up to 33% with topology-oblivious mapping and up to additional 7% with topology-aware mapping over the default sequential strategy.
准确和及时的天气现象预测，如飓风和山洪暴发，需要高保真的计算密集模拟多个精细区域的兴趣在一个粗略的模拟域。当前的天气应用程序使用所有可用的处理器依次执行这些嵌套模拟，由于其亚线性可扩展性，这不是最优的。在这项工作中，我们提出了一种并行执行多个嵌套域模拟的策略，该策略基于将二维处理器网格划分为与每个域相关联的不相交的矩形区域。我们提出了一种结合性能预测、处理器分配方法和环面互连区域拓扑感知映射的新方法。在IBM Blue Gene系统上使用WRF进行的实验表明，与默认顺序策略相比，所提出的策略在拓扑无关映射下的性能提高高达33%，在拓扑感知映射下的性能提高高达7%。
{"title":"A divide and conquer strategy for scaling weather simulations with multiple regions of interest","authors":"Preeti Malakar, Thomas George, Sameer Kumar, R. Mittal, V. Natarajan, Yogish Sabharwal, Vaibhav Saxena, Sathish S. Vadhiyar","doi":"10.3233/SPR-130367","DOIUrl":"https://doi.org/10.3233/SPR-130367","url":null,"abstract":"Accurate and timely prediction of weather phenomena, such as hurricanes and flash floods, require high-fidelity compute intensive simulations of multiple finer regions of interest within a coarse simulation domain. Current weather applications execute these nested simulations sequentially using all the available processors, which is sub-optimal due to their sub-linear scalability. In this work, we present a strategy for parallel execution of multiple nested domain simulations based on partitioning the 2-D processor grid into disjoint rectangular regions associated with each domain. We propose a novel combination of performance prediction, processor allocation methods and topology-aware mapping of the regions on torus interconnects. Experiments on IBM Blue Gene systems using WRF show that the proposed strategies result in performance improvement of up to 33% with topology-oblivious mapping and up to additional 7% with topology-aware mapping over the default sequential strategy.","PeriodicalId":217221,"journal":{"name":"2012 International Conference for High Performance Computing, Networking, Storage and Analysis","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129913020","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. Byna, J. Chou, O. Rübel, Prabhat, H. Karimabadi, W. Daughton, V. Roytershteyn, E. W. Bethel, Mark Howison, K. Hsu, Kuan-Wu Lin, A. Shoshani, A. Uselton, Kesheng Wu
Petascale plasma physics simulations have recently entered the regime of simulating trillions of particles. These unprecedented simulations generate massive amounts of data, posing significant challenges in storage, analysis, and visualization. In this paper, we present parallel I/O, analysis, and visualization results from a VPIC trillion particle simulation running on 120,000 cores, which produces ~30TB of data for a single timestep. We demonstrate the successful application of H5Part, a particle data extension of parallel HDF5, for writing the dataset at a significant fraction of system peak I/O rates. To enable efficient analysis, we develop hybrid parallel FastQuery to index and query data using multi-core CPUs on distributed memory hardware. We show good scalability results for the FastQuery implementation using up to 10,000 cores. Finally, we apply this indexing/query-driven approach to facilitate the first-ever analysis and visualization of the trillion particle dataset.
{"title":"Parallel I/O, analysis, and visualization of a trillion particle simulation","authors":"S. Byna, J. Chou, O. Rübel, Prabhat, H. Karimabadi, W. Daughton, V. Roytershteyn, E. W. Bethel, Mark Howison, K. Hsu, Kuan-Wu Lin, A. Shoshani, A. Uselton, Kesheng Wu","doi":"10.1109/SC.2012.92","DOIUrl":"https://doi.org/10.1109/SC.2012.92","url":null,"abstract":"Petascale plasma physics simulations have recently entered the regime of simulating trillions of particles. These unprecedented simulations generate massive amounts of data, posing significant challenges in storage, analysis, and visualization. In this paper, we present parallel I/O, analysis, and visualization results from a VPIC trillion particle simulation running on 120,000 cores, which produces ~30TB of data for a single timestep. We demonstrate the successful application of H5Part, a particle data extension of parallel HDF5, for writing the dataset at a significant fraction of system peak I/O rates. To enable efficient analysis, we develop hybrid parallel FastQuery to index and query data using multi-core CPUs on distributed memory hardware. We show good scalability results for the FastQuery implementation using up to 10,000 cores. Finally, we apply this indexing/query-driven approach to facilitate the first-ever analysis and visualization of the trillion particle dataset.","PeriodicalId":217221,"journal":{"name":"2012 International Conference for High Performance Computing, Networking, Storage and Analysis","volume":"130 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130420485","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}
Jinsuk Chung, Ikhwan Lee, Michael B. Sullivan, Jee Ho Ryoo, Dong-Wan Kim, D. Yoon, L. Kaplan, M. Erez
This paper describes and evaluates a scalable and efficient resilience scheme based on the concept of containment domains. Containment domains are a programming construct that enable applications to express resilience needs and to interact with the system to tune and specialize error detection, state preservation and restoration, and recovery schemes. Containment domains have weak transactional semantics and are nested to take advantage of the machine and application hierarchies and to enable hierarchical state preservation, restoration, and recovery. We evaluate the scalability and efficiency of containment domains using generalized trace-driven simulation and analytical analysis and show that containment domains are superior to both checkpoint restart and redundant execution approaches.
{"title":"Containment domains: A scalable, efficient, and flexible resilience scheme for exascale systems","authors":"Jinsuk Chung, Ikhwan Lee, Michael B. Sullivan, Jee Ho Ryoo, Dong-Wan Kim, D. Yoon, L. Kaplan, M. Erez","doi":"10.3233/SPR-130374","DOIUrl":"https://doi.org/10.3233/SPR-130374","url":null,"abstract":"This paper describes and evaluates a scalable and efficient resilience scheme based on the concept of containment domains. Containment domains are a programming construct that enable applications to express resilience needs and to interact with the system to tune and specialize error detection, state preservation and restoration, and recovery schemes. Containment domains have weak transactional semantics and are nested to take advantage of the machine and application hierarchies and to enable hierarchical state preservation, restoration, and recovery. We evaluate the scalability and efficiency of containment domains using generalized trace-driven simulation and analytical analysis and show that containment domains are superior to both checkpoint restart and redundant execution approaches.","PeriodicalId":217221,"journal":{"name":"2012 International Conference for High Performance Computing, Networking, Storage and Analysis","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114190420","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}
P. T. P. Tang, Jongsoo Park, Daehyun Kim, V. Petrov
In high-performance computing on distributed-memory systems, communication often represents a significant part of the overall execution time. The relative cost of communication will certainly continue to rise as compute-density growth follows the current technology and industry trends. Design of lower-communication alternatives to fundamental computational algorithms has become an important field of research. For distributed 1-D FFT, communication cost has hitherto remained high as all industry-standard implementations perform three all-to-all internode data exchanges (also called global transposes). These communication steps indeed dominate execution time. In this paper, we present a mathematical framework from which many single-all-to-all and easy-to-implement 1-D FFT algorithms can be derived. For large-scale problems, our implementation can be twice as fast as leading FFT libraries on state-of-the-art computer clusters. Moreover, our framework allows tradeoff between accuracy and performance, further boosting performance if reduced accuracy is acceptable.
{"title":"A framework for low-communication 1-D FFT","authors":"P. T. P. Tang, Jongsoo Park, Daehyun Kim, V. Petrov","doi":"10.3233/SPR-130373","DOIUrl":"https://doi.org/10.3233/SPR-130373","url":null,"abstract":"In high-performance computing on distributed-memory systems, communication often represents a significant part of the overall execution time. The relative cost of communication will certainly continue to rise as compute-density growth follows the current technology and industry trends. Design of lower-communication alternatives to fundamental computational algorithms has become an important field of research. For distributed 1-D FFT, communication cost has hitherto remained high as all industry-standard implementations perform three all-to-all internode data exchanges (also called global transposes). These communication steps indeed dominate execution time. In this paper, we present a mathematical framework from which many single-all-to-all and easy-to-implement 1-D FFT algorithms can be derived. For large-scale problems, our implementation can be twice as fast as leading FFT libraries on state-of-the-art computer clusters. Moreover, our framework allows tradeoff between accuracy and performance, further boosting performance if reduced accuracy is acceptable.","PeriodicalId":217221,"journal":{"name":"2012 International Conference for High Performance Computing, Networking, Storage and Analysis","volume":"199 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127946026","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}
A growing amount of data is produced daily resulting in a growing demand for storage solutions. While cloud storage providers offer a virtually infinite storage capacity, data owners seek geographical and provider diversity in data placement, in order to avoid vendor lock-in and to increase availability and durability. Moreover, depending on the customer data access pattern, a certain cloud provider may be cheaper than another. In this paper, we introduce Scalia, a cloud storage brokerage solution that continuously adapts the placement of data based on its access pattern and subject to optimization objectives, such as storage costs. Scalia efficiently considers repositioning of only selected objects that may significantly lower the storage cost. By extensive simulation experiments, we prove the cost-effectiveness of Scalia against static placements and its proximity to the ideal data placement in various scenarios of data access patterns, of available cloud storage solutions and of failures.
{"title":"Scalia: An adaptive scheme for efficient multi-cloud storage","authors":"Thanasis G. Papaioannou, N. Bonvin, K. Aberer","doi":"10.1109/SC.2012.101","DOIUrl":"https://doi.org/10.1109/SC.2012.101","url":null,"abstract":"A growing amount of data is produced daily resulting in a growing demand for storage solutions. While cloud storage providers offer a virtually infinite storage capacity, data owners seek geographical and provider diversity in data placement, in order to avoid vendor lock-in and to increase availability and durability. Moreover, depending on the customer data access pattern, a certain cloud provider may be cheaper than another. In this paper, we introduce Scalia, a cloud storage brokerage solution that continuously adapts the placement of data based on its access pattern and subject to optimization objectives, such as storage costs. Scalia efficiently considers repositioning of only selected objects that may significantly lower the storage cost. By extensive simulation experiments, we prove the cost-effectiveness of Scalia against static placements and its proximity to the ideal data placement in various scenarios of data access patterns, of available cloud storage solutions and of failures.","PeriodicalId":217221,"journal":{"name":"2012 International Conference for High Performance Computing, Networking, Storage and Analysis","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128430604","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：