– Major Greenplum 5.0 release planned early 2017 ## Greenplum Database Overview - Massively Parallel Processing (MPP) database system - Scales out to hundreds $ ^{(*)} $ of nodes • Shared nothing [Image](/uploads/documents/3/5/b/d/35bdf6490021398609e9a18998557eb1/p5_1.jpg) MPP Segment Host MPP Segment Host MPP Segment Host ## PostgreSQL Heritage ## PostgreSQL  Raster Image Processing  Spatial Indexes

Server 134 Query Processing 135 Data Flow 135 Cluster Management 136 Release 6.23 136 Release 6.23.3 136 Changed Feature 136 Resolved Issues 136 Extensions 136 Server 136 Query Processing 137 MADlib 138 138 Server 138 Query Processing 140 Data Flow 140 VMware Greenplum 6 Documentation VMware, Inc 3 Release 6.23.0 140 Features 140 Resolved Issues 142 Server 142 Query Processing 143 Cluster Management Release 6.22.2 144 Resolved Issues 144 Server 144 Query Processing 145 Data Flow 146 Release 6.22.1 146 Resolved Issues 146 Server 146 Query Processing 147 Cluster Management 147 Release 6.22.0 147 Features

Orientation 338 Compression 339 Distributions 339 Local (Co-located) Joins 340 Data Skew 340 Processing Skew 341 Partitioning 341 Number of Partition and Columnar Storage Files 342 Indexes 343 Column Backup and Restore Overview 0 Parallel Backup with gpbackup and gprestore 544 Non-Parallel Backup with pg_dump 544 Parallel Backup with gpbackup and gprestore 544 Parallel Backup with gpbackup and gprestore Viewing a Table's Distribution Key 616 Viewing Data Distribution 617 Checking for Query Processing Skew 617 Avoiding an Extreme Skew Warning 617 Viewing Metadata Information about Database Objects

New and Changed Features 143 Server 143 Query Processing 144 Release 6.27.0 144 Features 144 Resolved Issues 145 Server 145 Query Processing 146 Database Utilities 146 Cluster Management 146 Release 6.26.4 146 Server 147 Query Processing 147 Release 6.26.3 147 Server 147 Release 6.26.2 147 New Features 148 Query Processing 148 Server 148 Cluster Management 148 Disaster Supported Version 149 Query Processing 149 Server 149 Cluster Management 150 Release 6.26.0 150 Features 150 Resolved Issues 150 Server 151 Query Processing 151 Cluster Management 151

Orientation 333 Compression 334 Distributions 334 Local (Co-located) Joins 335 Data Skew 335 Processing Skew 336 Partitioning 336 Number of Partition and Columnar Storage Files 337 Indexes 338 Column Restoring Databases 536 Backup and Restore Overview 536 Parallel Backup with gpbackup and gprestore 536 Non-Parallel Backup with pg_dump 536 Parallel Backup with gpbackup and gprestore 537 Requirements 607 Viewing a Table’s Distribution Key 607 Viewing Data Distribution 607 Checking for Query Processing Skew 608 Avoiding an Extreme Skew Warning 608 Viewing Metadata Information about Database Objects

Release 6.22.2 134 Resolved Issues 134 Server 135 Query Processing 136 Data Flow 136 Release 6.22.1 136 Resolved Issues 136 Server 137 Query Processing 137 Cluster Management 138 Release 6.22.0 138 Features Features 138 Resolved Issues 140 Server 140 Query Processing 141 Data Flow 141 Cluster Management 141 DataDirect ODBC 141 Release 6.21 142 VMware Tanzu Greenplum 6 Documentation VMware, Inc. 3 Release 6.21 3 142 Server 142 Release 6.21.2 142 Server 142 Query Processing 142 Release 6.21.1 142 Features 142 Resolved Issues 142 Server 142 Query Processing 144 Data Flow 144 Tanzu Greenplum on vSphere 144 Release

mine! You can't have them. ## Agenda 1. Current status of C++ safety: MISRA and C++ CG 2. Parallel Safety rules 3. Automotive Safety case ## Safety Critical API Evolution  • WG23 (no parallel rules) • Added some from our own contributions • Many joined, average 5-8 per meeting

Agenda Greenplum简介 Greenplum的MPP架构 > 分布式优化器: Postgres planner 和 ORCA > 分布式事务和执行 Greenplum存储 Greenplum生态 Greenplum 7 Greenplum简介：什么是Greenplum? 基于PostgreSQL、开源、分布式MPP、ACID完备、为OLAP优化的关系型数据仓库。 https://greenplum 2003年，Luke Lonergan 和 Scott Yara 发起 Greenplum 项目，从 PostgreSQL 8 分支，做成 MPP 架构 2010年被EMC收购 2012年成为Pivotal的一部分 2015年开源，可能是世界上第一个成熟商用的开源MPP数据仓库 ➢ 2019年底跟随Pivotal被VMware收购 ## 谁在用Greenplum? ➢ 500多付费企业客户 成千上万的开源用户 616fd2ae0a0e1f69abbe60d5572e8/p5_11.jpg) ## Greenplum的MPP架构 ## Massively: ☐ PB级的数据，单台主机无法处理 ☐ 所以数据分布在多个主机上 ☐ 高效、灵活的数据分布，和实际业务相关 ## Parallel: ☐ 数据并行处理计算 ☐ 通过网络进行数据交换和汇总 ## 执行架构 ## ➤ Coordinator:

## Greenplum on Kubernetes 容器化MPP数据库 ## AGENDA ## 云数据库背景 云数据库实现方案 Greenplum on Kubernetes Greenplum Operator 总结 ## 云数据库背景 ## 云数据库背景 ● 资源变化 ☐ 本地资源 → 云 ☐ 静态资源 → 弹性需求 ## ● 数据变化 ☐ 内部数据 → 多数据源

## CurveBS I/O processing flow Before introducing IO processing flow, we first describe the overall architecture, data organization and topology structure of CURVE. ![Image](/uploads/documents/7/a/1 sockets. ● Nebdserver: Accepts requests from NEBDClient and calls Curve Client for corresponding processing. It can receive requests from different NEBDClients. 3. Through the above splitting, NebdClient NebdClient replaces Curve Client and directly interfaces with upper services. There is no logical processing in NEBDClient, it just proxy requests and has limited retries, which ensuring that NEBDClient needs no