https://github.com/jupyterlab/jupyterlab-desktop/releases jupyterlab桌面版，直接安装就可以用了。非常方便； Before installing please read the Python Environment Customization Guide if you plan to customize the Python environment to add new packages. If you have an existing JupyterLab Desktop installation, please uninstall it first by following the uninstall instructions. Debian, Ubuntu Linux Installer Red Hat, Fedora, SUSE Linux Installer macOS Installer Windows Installer Launching JupyterLab Desktop JupyterLab Desktop can be launched from the GUI of your operating system by clicking the application's icon or by using jlab command from the command line. Double clicking .ipynb files is also supported and it will launch JupyterLab Desktop and load the notebook file. JupyterLab Desktop sets File Browser's root directory based on the launch method. If launched from the application icon on GUI or by using jlab command without any arguments, then user's home is set as the root directory. If launched by double clicking .ipynb file or jlab command with a file path as the argument, then file's parent directory is set as the root directory. If jlab command is used with a directory path as the argument then the directory in the argument is set as the root directory.

jupyterlab免搭建安装包； https://github.com/jupyterlab/jupyterlab-desktop/releases 可在环境中直接进行包管理，非常方便，是windows和其他桌面环境用以演示项目的理想工具； 开发了pandas数据工具，图表渲染工具集，便于入门使用；

异构数据库对接，包括ClickHouse，Greenplum，postgreSQL等新型数据库产品，传统架构游刃有余； 1，不直接要求数据库有写权限，强调不通过存储过程和函数实现的ETL交互作业能力（全面在Kettle或者代码实现） 2，内存优化和SQL优化，避免锁库锁表，避免造成业务系统阻塞； 2-1,确认使用no_lock，放弃强一致性要求； 2-2,慢SQL查询准备； 2-3，表镜像，内存重构建业务； 2-4，动态判断，增量数据（not in）业务自动冗余； 3，全面构建日志，成功触发发送企业*；失败触发重动作（需运维介入）； 4，任务超时管理；