微软并行数据仓库技术概述在大数据存储中的应用解读.pptx

Microsoft Parallel Data Warehouse 微软并行数据仓库 2 Data sources 数据源 3 sources Data Non-Relational Data 数据源 非关系型数据 ? Data sources 数据源 Non-relational data 非关系型数据 Gain knowledge of all your existing data 通过任何数据获得知识 Enrich and optimize your data from non-traditional sources 通过非关系型数据源富集并优化数据 Keep legacy investment 保持原有投资 Buy new tier one hardware appliance 购买第一层硬件设备 Acquire big data solution 获取大数据解决方案 Acquire business intelligence 获取商业智能 Roadblocks to evolving to a modern data warehouse 现代化数据仓库面临的发展障碍 Limited scalability ability to handle new data types 扩展性有限，无法处理新型 数据 Significant training still siloed 需要大量培训，不同系 统相互隔离 High acquisition/ migration Costs 购买/迁移成本高 Complex with low adoption 过于复杂，接受度低 Microsoft Parallel Data Warehouse 微软并行数据仓库 The turnkey modern data warehouse appliance 现代化数据仓库设备一体机 “New “新”数据源 sources ” data Move HDFS into the warehouse before analysis 分析前需将 HDFS 迁移至仓库 ETL Learn new skills/ 学习新技能 TSQL Build Integrate Manage Maintain Support 构建集成式管理的 维护支持 Hadoop alone is not the answer to all big data challenges Hadoop 自身并非所有大数据问题的终极答案 Steep learning curve, slow and inefficient Hadoop ecosystem 陡峭的学习曲线，缓慢低效的 Hadoop 生态系统 “New” data sources “新”数据源 “New” data sources Single T-SQL query model for PDW and Hadoop with rich features of T-SQL including joins without ETL PDW 与 Hadoop 可使用通用的 T-SQL 查询模型，并具备丰富的 T-SQL 功能，包括无需 ETL 的连接 Leverages the power of MPP to enhance query execution performance 利用大规模并行处理的强大运算能力改善查询性能 Supports Windows Azure HDInsight to enable new hybrid cloud scenarios 支持 Windows Azure HDInsight，实现全新混合云场景 Query non-Microsoft Hadoop distributions such as Hortonworks and Cloudera 查询非微软 Hadoop 发行版，例如 Hortonworks 与 Cloudera SQL Server Parallel Data Warehouse SQL Server 并行数 据仓库 PolyBase Microsoft HDInsight Windows Azure HDInsight Cloudera Hortonworks Linux Select 选择… Result set 结果集 Connecting islands of data with PolyBase 使用 PolyBase 连接信息孤岛 Bringing Hadoop point solutions and the data wa