WBDB2012: Workshop on Big Data Benchmarking 2012

随着不同类型数据的规模、复杂性和获取速率的指数级增长，迫切需要用于管理和分析这些数据的新技术。在这方面，迫切需要制定基准，以促进对替代解决方案的评估，并对针对大数据应用的不同解决方案进行比较。基准测试需要捕捉大数据存储、管理和分析的各种特征，包括新的功能集、巨大的数据规模、大规模和不断发展的系统配置、不断变化的负载以及大数据和云平台的异构技术。这些基准不足以评估新出现的大数据平台、系统和软件，如SQL、NoSQL和Hadoop软件生态系统;不同形式或类型的大数据，包括图形、流、科学数据、文件集和交易数据;硬件方面的新选择，包括硬盘与固态硬盘、不同类型的硬盘、固态硬盘和主存以及大内存系统; 2012年大数据基准测试讲习班是朝着制定一套基准迈出的重要一步，这些基准可客观衡量处理大数据应用的硬件和软件系统的有效性。本次研讨会的目的是确定关键问题，并围绕能够捕捉大数据应用场景本质的参考基准定义开展活动。这项工作的目的是达到一套客观的措施和基准数据集，以表征和比较大数据存储，检索，处理和分析问题的替代解决方案的性能和价格/性能权衡。研讨会汇集了来自学术界和工业界的约40名专家，他们在大数据、数据库系统、基准测试和系统性能、云存储和计算以及相关领域具有背景。代表的行业包括硬件、软件、分析和应用。 该小组将编写一份报告草稿，介绍一套海量数据基准，并将在网上以及通过在有关会议和讲习班上的介绍和外联活动广泛传播。更广泛的影响：海量数据基准的提供将为比较关键海量数据问题的替代解决方案提供客观衡量标准，从而促进研究和技术进步。

With the exponential increase in the size, complexity, and rate of acquisition of diverse types of data, there an urgent need for new techniques for managing and analyzing such data. In this context, there is a critical need for benchmarks to facilitate evaluation of alternative solutions and provide for comparisons among different solution approaches targeted to big data applications. Benchmarks need to capture a variety of characteristics of big data storage, management, and analytics including new feature sets, enormous data size, largescale and evolving system configurations, shifting loads, and the heterogeneous technologies of big-data and cloud platforms. The benchmarks are inadequate for assessing emerging big data platforms, systems and in software such as SQL, NoSQL, and the Hadoop software ecosystem; different modalities or genres of big data, including graphs, streams, scientific data, document collections, and transaction data; new options in hardware including, HDD vs SSD, different types of HDD, SSD, and main memory, and large-memory systems; and, new platform options that include dedicated commodity clusters and cloud platforms.The Workshop on Big Data Benchmarking 2012 represents an important step towards the development of a suite of benchmarks for providing objective measures of the effectiveness of hardware and software systems dealing with big data applications. The objective of this invitation-only workshop is to identify key issues and to launch an activity around the definition of reference benchmarks that can capture the essence of big data application scenarios. The effort aims to arrive at a set of objective measures and benchmark datasets to characterize and compare the performance of and the price/performance tradeoffs of alternative solutions for big data storage, retrieval, processing, and analysis problems. The workshop brings together a group of about 40 experts from academia and industry with backgrounds in big data, database systems, benchmarking and system performance, cloud storage and computing, and related areas.The industries represented range from hardware, software, analytics, and applications. The group will develop a draft of a report describing a big data benchmark suite that will be widely disseminated on the web and through presentations and oureach activities at the relevant conferences and workshops.Broader Impacts: The availability of the big data benchmark suite will facilitate research and technological advances by providing objective measures for comparing alternative solutions to key big data problems.