The purpose of this special issue is to collect a selection of representative research articles that were primarily presented at the Eighth Workshop on Data-Intensive Computing in the Clouds, held in conjunction with SC'17. In particular, this annual workshop brings together domain scientists, researchers, scholars, vendors, and practitioners from the complementary fields of data science, cloud computing, and high-performance computing, in order to promote an exchange of ideas, discuss future collaborations, and develop new research directions. Data scientists increasingly rely on high-performance computers and cloud infrastructures to analyze high volumes of scientific data, automatically process data, and manage data privacy and performance. As scientific data continues to grow in volume and complexity, computational capabilities also increase at both supercomputing facilities and industry data center. Processing scientific data on emerging hardware with high performance-efficiency and privacy requires a knowledge combination from specific scientific domains and computer system techniques. This special issue presents examples of the successful collaboration from domain scientists, researchers, scholars, vendors and practitioners to address the research challenges on processing scientific data on the infrastructures of high-performance computers and cloud. The scope of this special issue is representative of the multidisciplinary nature of scientific computing in high-performance computing and cloud computing. This special issue addresses the challenges on practical experiences on processing scientific data in different domains and different platforms in high-performance computers and cloud. In particular, Zamani et al.1 show how to automatically integrate large-scale facilities with cyberinfrastructure services for automated data processing. Peng and Plale2 identify the requirements for managing computational analysis among candidate storage solutions. Koulouzis et al.3 describe their experiences that identify the time-critical requirements of environmental scientists making use of computational research support environments and provide a case study whereby their software suite is used to optimize runtime service quality for a data subscription service. Dayarathna and Suzumura4 demonstrate their approach on producing optimized stream query performance, and further compare the solution to naive deployments using two real-world stream processing applications in the domains of health care and search advertising. We encourage the readers to review the aforementioned articles to gain insight into the breadth and depth of problems and innovative solutions in the multidisciplinary field of scientific data in high-performance computing and cloud computing.

中文翻译：

---

云中数据密集型计算研讨会特刊前言

本期特刊的目的是收集精选的代表性研究文章，这些文章主要在与 SC'17 联合举办的第八届云中数据密集型计算研讨会上发表。特别是，这个年度研讨会汇集了来自数据科学、云计算和高性能计算等互补领域的领域科学家、研究人员、学者、供应商和从业者，以促进思想交流、讨论未来合作以及开拓新的研究方向。数据科学家越来越依赖高性能计算机和云基础设施来分析大量科学数据、自动处理数据以及管理数据隐私和性能。随着科学数据的数量和复杂性不断增加，超级计算设施和工业数据中心的计算能力也在增加。在具有高性能和隐私性的新兴硬件上处理科学数据需要来自特定科学领域和计算机系统技术的知识组合。本期特刊介绍了领域科学家、研究人员、学者、供应商和从业人员成功合作的例子，以解决在高性能计算机和云基础设施上处理科学数据的研究挑战。本期特刊的范围代表了高性能计算和云计算中科学计算的多学科性质。本期特刊解决了在高性能计算机和云中处理不同领域和不同平台科学数据的实践经验的挑战。特别是，Zamani 等人 1 展示了如何将大型设施与网络基础设施服务自动集成以进行自动化数据处理。Peng 和 Plale2 确定了在候选存储解决方案之间管理计算分析的要求。Kolouzis 等人 3 描述了他们的经验，这些经验确定了环境科学家利用计算研究支持环境的时间关键要求，并提供了一个案例研究，其中他们的软件套件用于优化数据订阅服务的运行时服务质量。Dayarathna 和 Suzumura4 展示了他们产生优化流查询性能的方法，并在医疗保健和搜索广告领域使用两个真实世界的流处理应用程序，进一步将解决方案与简单部署进行比较。我们鼓励读者阅读上述文章，以深入了解高性能计算和云计算中科学数据多学科领域中问题的广度和深度以及创新解决方案。