<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[binli6949的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/binli6949</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; binli6949]]></copyright><item><title><![CDATA[基于flink的协同过滤]]></title><link>https://blog.youkuaiyun.com/binli6949/article/details/100950742</link><guid>https://blog.youkuaiyun.com/binli6949/article/details/100950742</guid><author>binli6949</author><pubDate>Fri, 30 Aug 2019 11:46:27 +0800</pubDate><description><![CDATA[最近flink较火，尝试使用flink做推荐功能试试，说干就干，话说flink-ml确实比较水，包含的算法较少，且只支持scala版本，以至flink1.9已经将flink-ml移除，看来是准备有大动作，但后期的实时推荐，flink能派上大用场。所幸基于物品的协同过滤算法相对简单，实现起来难度不大。先看目前推荐整体的架构。

先说一下用到的相似算法：X=（x1, x2, x3, … xn），Y=（...]]></description><category></category></item><item><title><![CDATA[flink 扩容]]></title><link>https://blog.youkuaiyun.com/binli6949/article/details/100950753</link><guid>https://blog.youkuaiyun.com/binli6949/article/details/100950753</guid><author>binli6949</author><pubDate>Tue, 06 Aug 2019 10:07:32 +0800</pubDate><description><![CDATA[因需要，再扩3台flink， 遇到各种报错，折腾了3个小时，最终找到了一种简单可行的方式，该方式，需要清空checkpoint， 暂时没有找到更好的方案。
rm -rf /tmprm -rf /data/hadoop/*  （checkpoint存储的位置）hadoop namenode -format(所有结点)start-dfs.sh start-yarn.shstart-zookeeper-...]]></description><category></category></item><item><title><![CDATA[flink 优秀文章集合]]></title><link>https://blog.youkuaiyun.com/binli6949/article/details/100950744</link><guid>https://blog.youkuaiyun.com/binli6949/article/details/100950744</guid><author>binli6949</author><pubDate>Tue, 09 Jul 2019 19:51:15 +0800</pubDate><description><![CDATA[说明：为了避免博客被封，*代表c,s,d和n,请自行替换.  

水印

https://blog.****.net/lmalds/article/details/52704170
2  python 操作
https://blog.****.net/qq_41577045/article/details/86694395...]]></description><category></category></item><item><title><![CDATA[利用flink统计消息回复情况]]></title><link>https://blog.youkuaiyun.com/binli6949/article/details/100950751</link><guid>https://blog.youkuaiyun.com/binli6949/article/details/100950751</guid><author>binli6949</author><pubDate>Tue, 09 Jul 2019 14:23:44 +0800</pubDate><description><![CDATA[其中用到了滑动窗口函数大小30秒，间隔15秒，且大于窗口10秒的数据，被丢弃。（实际业务这三个值 应为是 10 分钟，1分钟，5分钟）。代码先记录一下
public static void main(String[] arg) throws Exception {

        final StreamExecutionEnvironment env = StreamExecutionEnvi...]]></description><category></category></item><item><title><![CDATA[flink  sql-clent  MATCH_RECOGNIZE kafka 例子]]></title><link>https://blog.youkuaiyun.com/binli6949/article/details/100950748</link><guid>https://blog.youkuaiyun.com/binli6949/article/details/100950748</guid><author>binli6949</author><pubDate>Fri, 05 Jul 2019 14:05:29 +0800</pubDate><description><![CDATA[环境  flink1.7.2


增加flink1.7.2 的lib 中的jar， 否则会报类找不到
avro-1.8.2.jar            
flink-connector-kafka-0.10_2.12-1.7.2.jar  
flink-connector-kafka-base_2.12-1.7.2.jar  
flink-json-1.7.2.jar         
kafk...]]></description><category></category></item></channel></rss>