SpringBoot集成Mongodb
1. SpringBoot集成
pom.xml
1 | <dependency> |
application.properties
1 | =mongodb://user:pwd@ip:27017/database |
集群版mongodb
1 | =mongodb://user:pwd@ip1:port1,ip2:port2/database |
pom.xml
1 | <dependency> |
application.properties
1 | =mongodb://user:pwd@ip:27017/database |
集群版mongodb
1 | =mongodb://user:pwd@ip1:port1,ip2:port2/database |
Why Scroll?
Scrolling is the fundamental interaction of the Internet. Having animated elements triggered on scroll is a common practice, but what if you want to take it to another level? For example, websites like r2d3 — A visual introduction to machine learning and NASA: Prospect all features full-scale animations that users can control and interact with through scrolling.
I was really drawn to this form of interaction, which has inspired me to create OKAI — An Interactive Introduction to Artificial Intelligence(AI). Every chapter (except Chapter 0) features full-scale scroll-based animations that visualize and break down complicated concepts in AI. Allowing users to control the pace of the animations give them a sense of control and security when exploring the unknown world of AI principles.
https://okai.brown.edu/chapter1.html
In this article, I am presenting the workflow of how these animations are created and controlled through scrolling.
最近新购入一块3T硬盘,准备划给Linux 系统2T,用来玩数据分析。开始新硬盘的挂载之旅
1 | [root@sloong sloong]# fdisk -l |
可以看到新磁盘为 /dev/sda, 3000G, 扇区 0.5k/4k ,IO大小 4k/4k
1 | hdfs文件的相关操作主要使用hadoop fs、hadoop dfs、hdfs dfs 命令,以下对最常用的相关命令进行简要说明。 |
背景:用户界面中,为了改善用户使用体验,移除了 查询时点击按钮的操作,变更为只要检测到查询条件的修改都会自动触发计算。而实际使用过程中,用户在最终条件确定前,所有条件变更导致的查询计算均是计算资源的浪费
目的:为了避免自动触发的计算导致Presto 计算资源的浪费

如图所示,左侧指标、细分维度、公共过滤条件以及 日期范围、日期粒度、人群的变化都会导致分析查询的调用
方案:
HBase由于其存储和读写的高性能,在OLAP即时分析中越来越发挥重要的作用,在易观精细化运营产品–易观方舟也有广泛的应用。作为Nosql数据库的一员,HBase查询只能通过其Rowkey来查询(Rowkey用来表示唯一一行记录),Rowkey设计的优劣直接影响读写性能。HBase中的数据是按照Rowkey的ASCII字典顺序进行全局排序的,有伙伴可能对ASCII字典序印象不够深刻,下面举例说明:
假如有5个Rowkey:”012”, “0”, “123”, “234”, “3”,按ASCII字典排序后的结果为:”0”, “012”, “123”, “234”, “3”。(注:文末附常用ASCII码表)
Rowkey排序时会先比对两个Rowkey的第一个字节,如果相同,然后会比对第二个字节,依次类推… 对比到第X个字节时,已经超出了其中一个Rowkey的长度,短的Rowkey排在前面。
由于HBase是通过Rowkey查询的,一般Rowkey上都会存一些比较关键的检索信息,我们需要提前想好数据具体需要如何查询,根据查询方式进行数据存储格式的设计,要避免做全表扫描,因为效率特别低。

Update your browser to view this website correctly. Update my browser now