Zhuo Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Zhuo Liu has a total of 57 co-authors. Here are the top co-authors by number of shared publications:
Weikuan Yu
12 shared publicationsBin Wang
6 shared publicationsTeng Wang
Florida State University, Tallahassee, FL, USA
5 shared publicationsCong Xu
Auburn University, Auburn, AL, USA
4 shared publicationsBin Wang
3 shared publications- Fang-Jian Zhou
Department of Urology, Sun Yat-Sen University Cancer Center, Guangzhou, China
3 shared publications
Latest Publications
- Top Cited Papers
- 1 May 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1789-1792
Benchmarking Streaming Computation Engines: Storm, Flink and Spark Streaming
Streaming data processing has been gaining attention due to its application into a wide range of scenarios. To serve the booming demands of streaming data processing, many computation engines have been developed. However, there is still a lack of real-world benchmarks that would be helpful when choosing the most appropriate platform for serving real-time streaming needs. In order to address this problem, we developed a streaming benchmark for three representative computation engines: Flink, Storm and Spark Streaming. Instead of testing speed-of-light event processing, we construct a full data pipeline using Kafka and Redis in order to more closely mimic the real-world production scenarios. Based on our experiments, we provide a performance comparison of the three data engines in terms of 99th percentile latency and throughput for various configurations.
- 1 May 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1789-1792
Top Cited PapersBenchmarking Streaming Computation Engines: Storm, Flink and Spark Streaming
Streaming data processing has been gaining attention due to its application into a wide range of scenarios. To serve the booming demands of streaming data processing, many computation engines have been developed. However, there is still a lack of real-world benchmarks that would be helpful when choosing the most appropriate platform for serving real-time streaming needs. In order to address this problem, we developed a streaming benchmark for three representative computation engines: Flink, Storm and Spark Streaming. Instead of testing speed-of-light event processing, we construct a full data pipeline using Kafka and Redis in order to more closely mimic the real-world production scenarios. Based on our experiments, we provide a performance comparison of the three data engines in terms of 99th percentile latency and throughput for various configurations.
- 3 December 2015
- Published by SAGE Publications in The International Journal of High Performance Computing Applications
- Vol. 30 (2) , 241-256
Enhance parallel input/output with cross-bundle aggregation
The exponential growth of computing power on leadership scale computing platforms imposes grand challenge to scientific applications’ input/output (I/O) performance. To bridge the performance gap between computation and I/O, various parallel I/O libraries have been developed and adopted by computer scientists. These libraries enhance the I/O parallelism by allowing multiple processes to concurrently access the shared data set. Meanwhile, they are integrated with a set of I/O optimization strategies such as data sieving and two-phase I/O to better exploit the supplied bandwidth of the underlying parallel file system. Most of these techniques are optimized for the access on a single bundle of variables generated by the scientific applications during the I/O phase, which is stored in the form of file. Few of these techniques focus on cross-bundle I/O optimizations. In this article, we investigate the potential benefit from cross-bundle I/O aggregation. Based on the analysis of the I/O patterns of a mission-critical scientific application named the Goddard Earth Observing System, version 5 (GEOS-5), we propose a Bundle-based PARallel Aggregation (BPAR) framework with three partitioning schemes to improve its I/O performance as well as the I/O performance of a broad range of other scientific applications. Our experiment result reveals that BPAR can deliver 2.1× I/O performance improvement over the baseline GEOS-5, and it is very promising in accelerating scientific applications’ I/O performance on various computing platforms.
- 3 December 2015
- Published by SAGE Publications in The International Journal of High Performance Computing Applications
- Vol. 30 (2) , 241-256
Enhance parallel input/output with cross-bundle aggregation
The exponential growth of computing power on leadership scale computing platforms imposes grand challenge to scientific applications’ input/output (I/O) performance. To bridge the performance gap between computation and I/O, various parallel I/O libraries have been developed and adopted by computer scientists. These libraries enhance the I/O parallelism by allowing multiple processes to concurrently access the shared data set. Meanwhile, they are integrated with a set of I/O optimization strategies such as data sieving and two-phase I/O to better exploit the supplied bandwidth of the underlying parallel file system. Most of these techniques are optimized for the access on a single bundle of variables generated by the scientific applications during the I/O phase, which is stored in the form of file. Few of these techniques focus on cross-bundle I/O optimizations. In this article, we investigate the potential benefit from cross-bundle I/O aggregation. Based on the analysis of the I/O patterns of a mission-critical scientific application named the Goddard Earth Observing System, version 5 (GEOS-5), we propose a Bundle-based PARallel Aggregation (BPAR) framework with three partitioning schemes to improve its I/O performance as well as the I/O performance of a broad range of other scientific applications. Our experiment result reveals that BPAR can deliver 2.1× I/O performance improvement over the baseline GEOS-5, and it is very promising in accelerating scientific applications’ I/O performance on various computing platforms.
- 19 November 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15267539,p. 198-207
Preserving Row Buffer Locality for PCM Wear-Leveling under Massive Parallelism
Phase Change Memory (PCM) is a promising alternative technology for DRAM because of its advantages in terms of transistor density and energy consumption. It has been exploited to work in concert or alone inside various memory systems to meet the growing bandwidth needs of massive parallelism. PCM memory cells, however, have a common problem of limited write endurance. Various wear-leaving techniques have been employed for uniform distribution of memory writes, typically through address transformation schemes such as randomization to avoid hot writes. Unfortunately, such address transformation can have the undesirable consequence of disrupting the row buffer locality in sequential memory accesses, resulting in the loss of memory performance. Our analysis reveals that this situation is particularly severe under massive parallelism of manycore processors such as GPUs. In this paper, we introduce a combination of two techniques, matrix-based partial randomization and rowbuffer locality-aware rotation, to alleviate the locality disruption of address transformation and preserve the row buffer locality of PCM-based global memory in GPU. Our evaluation results show that, compared to existing techniques, our techniques can adequately preserve the row buffer locality and minimize the loss of memory performance, while achieving similar endurance and better energy efficiency for a variety of GPGPU applications.
- 19 November 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15267539,p. 198-207
Preserving Row Buffer Locality for PCM Wear-Leveling under Massive Parallelism
Phase Change Memory (PCM) is a promising alternative technology for DRAM because of its advantages in terms of transistor density and energy consumption. It has been exploited to work in concert or alone inside various memory systems to meet the growing bandwidth needs of massive parallelism. PCM memory cells, however, have a common problem of limited write endurance. Various wear-leaving techniques have been employed for uniform distribution of memory writes, typically through address transformation schemes such as randomization to avoid hot writes. Unfortunately, such address transformation can have the undesirable consequence of disrupting the row buffer locality in sequential memory accesses, resulting in the loss of memory performance. Our analysis reveals that this situation is particularly severe under massive parallelism of manycore processors such as GPUs. In this paper, we introduce a combination of two techniques, matrix-based partial randomization and rowbuffer locality-aware rotation, to alleviate the locality disruption of address transformation and preserve the row buffer locality of PCM-based global memory in GPU. Our evaluation results show that, compared to existing techniques, our techniques can adequately preserve the row buffer locality and minimize the loss of memory performance, while achieving similar endurance and better energy efficiency for a variety of GPGPU applications.
- 7 January 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 27 (1) , 185-196
Exploiting Analytics Shipping with Virtualized MapReduce on HPC Backend Storage Servers
Large-scale scientific applications on High-Performance Computing (HPC) systems are generating a colossal amount of data that need to be analyzed in a timely manner for new knowledge, but are too costly to transfer due to their sheer size. Many HPC systems have catered to in situ analytics solutions that can analyze temporary datasets as they are generated, i.e., without storing to long-term storage media. However, there is still an open question on how to conduct efficient analytics of permanent datasets that have been stored to the backend persistent storage because of their long-term value. To fill the void, we exploit the analytics shipping model for fast analysis of large-scale scientific datasets on HPC backend storage servers. Through an efficient integration of MapReduce and the popular Lustre storage system, we have developed a Virtualized Analytics Shipping (VAS) framework that can ship MapReduce programs to Lustre storage servers. The VAS framework includes three component techniques: (a) virtualized analytics shipping with fast network and disk I/O; (b) stripe-aligned data distribution and task scheduling and (c) pipelined intermediate data merging and reducing. The first technique provides necessary isolation between MapReduce analytics and Lustre I/O services. The second and third techniques optimize MapReduce on Lustre and avoid explicit shuffling. Our performance evaluation demonstrates that VAS offers an exemplary implementation of analytics shipping and delivers fast and virtualized MapReduce programs on backend Lustre storage servers.
- 7 January 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 27 (1) , 185-196
Exploiting Analytics Shipping with Virtualized MapReduce on HPC Backend Storage Servers
Large-scale scientific applications on High-Performance Computing (HPC) systems are generating a colossal amount of data that need to be analyzed in a timely manner for new knowledge, but are too costly to transfer due to their sheer size. Many HPC systems have catered to in situ analytics solutions that can analyze temporary datasets as they are generated, i.e., without storing to long-term storage media. However, there is still an open question on how to conduct efficient analytics of permanent datasets that have been stored to the backend persistent storage because of their long-term value. To fill the void, we exploit the analytics shipping model for fast analysis of large-scale scientific datasets on HPC backend storage servers. Through an efficient integration of MapReduce and the popular Lustre storage system, we have developed a Virtualized Analytics Shipping (VAS) framework that can ship MapReduce programs to Lustre storage servers. The VAS framework includes three component techniques: (a) virtualized analytics shipping with fast network and disk I/O; (b) stripe-aligned data distribution and task scheduling and (c) pipelined intermediate data merging and reducing. The first technique provides necessary isolation between MapReduce analytics and Lustre I/O services. The second and third techniques optimize MapReduce on Lustre and avoid explicit shuffling. Our performance evaluation demonstrates that VAS offers an exemplary implementation of analytics shipping and delivers fast and virtualized MapReduce programs on backend Lustre storage servers.
- 1 January 2015
- Published by EDAA
- p. 689-694
Eliminating Intra-Warp Conflict Misses in GPU
Cache indexing functions play a key role in reducing conflict misses by spreading accesses evenly among all sets of cache blocks. Although various methods have been proposed, no significant effort has been expended on the behavior of conflict misses ...
- 1 January 2015
- Published by EDAA
- p. 689-694
Eliminating Intra-Warp Conflict Misses in GPU
Cache indexing functions play a key role in reducing conflict misses by spreading accesses evenly among all sets of cache blocks. Although various methods have been proposed, no significant effort has been expended on the behavior of conflict misses ...
- 1 November 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 25-32
BPAR: A Bundle-Based Parallel Aggregation Framework for Decoupled I/O Execution
In today's "Big Data" era, developers have adopted I/O techniques such as MPI-IO, Parallel NetCDF and HDF5 to garner enough performance to manage the vast amount of data that scientific applications require. These I/O techniques offer parallel access to shared datasets and together with a set of optimizations such as data sieving and two-phase I/O to boost I/O throughput. While most of these techniques focus on optimizing the access pattern on a single file or file extent, few of these techniques consider cross-file I/O optimizations. This paper aims to explore the potential benefit from cross-file I/O aggregation. We propose a Bundle-based PARallel Aggregation framework (BPAR) and design three partitioning schemes under such framework that targets at improving the I/O performance of a mission-critical application GEOS-5, as well as a broad range of other scientific applications. The results of our experiments reveal that BPAR can achieve on average 2.1× performance improvement over the baseline GEOS-5.
- 1 November 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 25-32
BPAR: A Bundle-Based Parallel Aggregation Framework for Decoupled I/O Execution
In today's "Big Data" era, developers have adopted I/O techniques such as MPI-IO, Parallel NetCDF and HDF5 to garner enough performance to manage the vast amount of data that scientific applications require. These I/O techniques offer parallel access to shared datasets and together with a set of optimizations such as data sieving and two-phase I/O to boost I/O throughput. While most of these techniques focus on optimizing the access pattern on a single file or file extent, few of these techniques consider cross-file I/O optimizations. This paper aims to explore the potential benefit from cross-file I/O aggregation. We propose a Bundle-based PARallel Aggregation framework (BPAR) and design three partitioning schemes under such framework that targets at improving the I/O performance of a mission-critical application GEOS-5, as well as a broad range of other scientific applications. The results of our experiments reveal that BPAR can achieve on average 2.1× performance improvement over the baseline GEOS-5.
- 1 January 2014
- Published by Springer Nature in Lecture Notes in Computer Science
- p. 148-163
Assessing the Performance Impact of High-Speed Interconnects on MapReduce
- 1 January 2014
- Published by Springer Nature in Lecture Notes in Computer Science
- p. 148-163
Assessing the Performance Impact of High-Speed Interconnects on MapReduce
- 1 September 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15525244,p. 1-8
A case of system-wide power management for scientific applications
The advance of high-performance computing systems towards exascale will be constrained by the systems' energy consumption levels. Large numbers of processing components, memory, interconnects, and storage components must all be considered to achieve exascale performance within a targeted energy bound. While application-aware power allocation schemes for computing resources are well studied, a portable and scalable budget-constrained power management scheme for scientific applications on exascale systems is still required. Execution activities within scientific applications can be categorized as CPU-bound, I/O-bound and communication-bound. Such activities tend to be clustered into `phases', offering opportunities to manage their power consumption separately. Our experiments have demonstrated that their performance and energy consumption are affected differently by CPU frequency, an opportunity to fine tune CPU frequency for a minimal impact on the total execution time but significant savings on the energy consumption. By exploiting this opportunity, we present a phase-aware hierarchical power management framework that can opportunistically deliver good tradeoffs between system power consumption and application performance under a power budget. Our hierarchical power management framework consists of two main techniques: Phase-Aware CPU Frequency Scaling (PAFS) and opportunistic provisioning for power-constrained performance optimization. We have performed a systematic evaluation using both simulations and representative scientific applications on real systems. Our results show that our techniques can achieve 4.3%-17% better energy efficiency for large-scale scientific applications.
- 1 September 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15525244,p. 1-8
A case of system-wide power management for scientific applications
The advance of high-performance computing systems towards exascale will be constrained by the systems' energy consumption levels. Large numbers of processing components, memory, interconnects, and storage components must all be considered to achieve exascale performance within a targeted energy bound. While application-aware power allocation schemes for computing resources are well studied, a portable and scalable budget-constrained power management scheme for scientific applications on exascale systems is still required. Execution activities within scientific applications can be categorized as CPU-bound, I/O-bound and communication-bound. Such activities tend to be clustered into `phases', offering opportunities to manage their power consumption separately. Our experiments have demonstrated that their performance and energy consumption are affected differently by CPU frequency, an opportunity to fine tune CPU frequency for a minimal impact on the total execution time but significant savings on the energy consumption. By exploiting this opportunity, we present a phase-aware hierarchical power management framework that can opportunistically deliver good tradeoffs between system power consumption and application performance under a power budget. Our hierarchical power management framework consists of two main techniques: Phase-Aware CPU Frequency Scaling (PAFS) and opportunistic provisioning for power-constrained performance optimization. We have performed a systematic evaluation using both simulations and representative scientific applications on real systems. Our results show that our techniques can achieve 4.3%-17% better energy efficiency for large-scale scientific applications.
- 1 July 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10952055,p. 1-7
Profiling and Improving I/O Performance of a Large-Scale Climate Scientific Application
Exascale computing systems are soon to emerge, which will pose great challenges on the huge gap between computing and I/O performance. Many large-scale scientific applications play an important role in our daily life. The huge amounts of data generated by such applications require highly parallel and efficient I/O management policies. In this paper, we adopt a mission-critical scientific application, GEOS-5, as a case to profile and analyze the communication and I/O issues that are preventing applications from fully utilizing the underlying parallel storage systems. Through in-detail architectural and experimental characterization, we observe that current legacy I/O schemes incur significant network communication overheads and are unable to fully parallelize the data access, thus degrading applications' I/O performance and scalability. To address these inefficiencies, we redesign its I/O framework along with a set of parallel I/O techniques to achieve high scalability and performance. Evaluation results on the NASA discover cluster show that our optimization of GEOS- 5 with ADIOS has led to significant performance improvements compared to the original GEOS-5 implementation.
- 1 July 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10952055,p. 1-7
Profiling and Improving I/O Performance of a Large-Scale Climate Scientific Application
Exascale computing systems are soon to emerge, which will pose great challenges on the huge gap between computing and I/O performance. Many large-scale scientific applications play an important role in our daily life. The huge amounts of data generated by such applications require highly parallel and efficient I/O management policies. In this paper, we adopt a mission-critical scientific application, GEOS-5, as a case to profile and analyze the communication and I/O issues that are preventing applications from fully utilizing the underlying parallel storage systems. Through in-detail architectural and experimental characterization, we observe that current legacy I/O schemes incur significant network communication overheads and are unable to fully parallelize the data access, thus degrading applications' I/O performance and scalability. To address these inefficiencies, we redesign its I/O framework along with a set of parallel I/O techniques to achieve high scalability and performance. Evaluation results on the NASA discover cluster show that our optimization of GEOS- 5 with ADIOS has led to significant performance improvements compared to the original GEOS-5 implementation.
- 1 May 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-10
A lightweight I/O scheme to facilitate spatial and temporal queries of scientific data analytics
In the era of petascale computing, more scientific applications are being deployed on leadership scale computing platforms to enhance the scientific productivity. Many I/O techniques have been designed to address the growing I/O bottleneck on large-scale systems by handling massive scientific data in a holistic manner. While such techniques have been leveraged in a wide range of applications, they have not been shown as adequate for many mission critical applications, particularly in data postprocessing stage. One of the examples is that some scientific applications generate datasets composed of a vast amount of small data elements that are organized along many spatial and temporal dimensions but require sophisticated data analytics on one or more dimensions. Including such dimensional knowledge into data organization can be beneficial to the efficiency of data post-processing, which is often missing from exiting I/O techniques. In this study, we propose a novel I/O scheme named STAR (Spatial and Temporal AggRegation) to enable high performance data queries for scientific analytics. STAR is able to dive into the massive data, identify the spatial and temporal relationships among data variables, and accordingly organize them into an optimized multi-dimensional data structure before storing to the storage. This technique not only facilitates the common access patterns of data analytics, but also further reduces the application turnaround time. In particular, STAR is able to enable efficient data queries along the time dimension, a practice common in scientific analytics but not yet supported by existing I/O techniques. In our case study with a critical climate modeling application GEOS-5, the experimental results on Jaguar supercomputer demonstrate an improvement up to 73 times for the read performance compared to the original I/O method.
- 1 May 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-10
A lightweight I/O scheme to facilitate spatial and temporal queries of scientific data analytics
In the era of petascale computing, more scientific applications are being deployed on leadership scale computing platforms to enhance the scientific productivity. Many I/O techniques have been designed to address the growing I/O bottleneck on large-scale systems by handling massive scientific data in a holistic manner. While such techniques have been leveraged in a wide range of applications, they have not been shown as adequate for many mission critical applications, particularly in data postprocessing stage. One of the examples is that some scientific applications generate datasets composed of a vast amount of small data elements that are organized along many spatial and temporal dimensions but require sophisticated data analytics on one or more dimensions. Including such dimensional knowledge into data organization can be beneficial to the efficiency of data post-processing, which is often missing from exiting I/O techniques. In this study, we propose a novel I/O scheme named STAR (Spatial and Temporal AggRegation) to enable high performance data queries for scientific analytics. STAR is able to dive into the massive data, identify the spatial and temporal relationships among data variables, and accordingly organize them into an optimized multi-dimensional data structure before storing to the storage. This technique not only facilitates the common access patterns of data analytics, but also further reduces the application turnaround time. In particular, STAR is able to enable efficient data queries along the time dimension, a practice common in scientific analytics but not yet supported by existing I/O techniques. In our case study with a critical climate modeling application GEOS-5, the experimental results on Jaguar supercomputer demonstrate an improvement up to 73 times for the read performance compared to the original I/O method.