Intelligent Placement of Apps and Users on Qlik Analytics Engines

Qlik Analytics Engine 上的应用程序和用户的智能放置

• 批准号: 520265-2017
• 负责人: Viktor, Herna
• 金额: $ 1.82万
• 依托单位: University of Ottawa
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2017
• 资助国家: 加拿大
• 起止时间: 2017-01-01 至 2018-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=640138
• 关键词: Intelligent; Placement; Apps; Users; Qlik

Business analytics tools provide intuitive, visual analytics solutions to allow organizations to discover the storythat lives within their data. Increasingly, business intelligence solutions are utilized across a variety ofindustries, including, healthcare, financial services, retail and life sciences. In this setting, there is a need formachine learning solutions that may be used to optimize the resource-usage of analytics applications incloud-based systems. There are several factors which need to be taken into consideration when aiming to findthe optimal intelligent analytics application placement solutions. Firstly, the applications' behaviours inmemory, and the CPU usages, are determined by a number of factors such as the amount of data included, thecardinality of the data, the complexity of the operations (e.g. aggregation or extensive sorting), the nature of thedata model, as well as the sizes and locations of the data sources. In addition, the usage patterns of differenttypes of users may vary considerably. For instance, push-button knowledge workers' usages may be far lessresource-intensive than that of data scientists, who would typically perform advanced, ad hoc analytics. Thisproject concerns the development of machine intelligence algorithms to address these issues, by buildingadaptive solutions to greatly improve placement policies of analytics applications across cloud-based platforms.

业务分析工具提供直观、可视化的分析解决方案，让组织能够发现数据中的故事。商业智能解决方案越来越多地用于各种行业，包括医疗保健、金融服务、零售和生命科学。在这种情况下，需要一种机器学习解决方案，可以用于优化基于计算机的系统的分析应用程序的资源使用。在寻找最佳智能分析应用程序放置解决方案时，需要考虑几个因素。首先，应用程序在内存中的行为和CPU使用率由许多因素决定，例如包含的数据量，数据的基数，操作的复杂性（例如聚合或广泛排序），数据模型的性质以及数据源的大小和位置。此外，不同类型用户的使用模式可能差异很大。例如，按钮式知识工作者的使用可能比数据科学家的使用少得多，数据科学家通常会执行高级的临时分析。该项目关注机器智能算法的开发，以解决这些问题，通过构建自适应解决方案，大大提高跨云平台的分析应用程序的放置策略。