ºÝºÝߣ

ºÝºÝߣShare a Scribd company logo
Personal Information
Organization / Workplace
China China
Occupation
Engineer / Big Data Arch / Hadoop / Spark / HBase / Kafka / Flume
Industry
Technology / Software / Internet
About
Taobao's hadoop cluster started at 2008,when the alibaba group started to realize that the big data might be very important in the future. The team established in 2008 and Li is the first batch of employees. The hadoop cluster named as Yunti,which means scaling ladder in English. The Yunti hadoop cluster started with the hadoop version of 0.19.1 and the scale was 200 nodes at the beginning in 2008. Five years later, in 2013, the cluster's scale is now 10000 nodes with 200PB data after compressed and erasure coded. The Version of yunti is still 0.19.1, but with about 800 patches on it in the svn of taobao. Li is the leader of the hadoop development team and hands-on improve most the ha...
Contact Details

Presentations(4)?

See all
The application and architecture of the real-time computing platform in di di-flink_forward_2018
The application and architecture of the real-time computing platform in di di-flink_forward_2018The application and architecture of the real-time computing platform in di di-flink_forward_2018
The application and architecture of the real-time computing platform in di di-flink_forward_2018
?
Qcon2013 ÂÞÀî - hadoopÔÚ°¢Àï
Qcon2013 ÂÞÀî - hadoopÔÚ°¢ÀïQcon2013 ÂÞÀî - hadoopÔÚ°¢Àï
Qcon2013 ÂÞÀî - hadoopÔÚ°¢Àï
?
±á»å´Ú²õÔªÊý¾ÝµÄ¶ÀÁ¢·þÎñºÍ¶ÀÁ¢³Ö¾Ã»¯´æ´¢
±á»å´Ú²õÔªÊý¾ÝµÄ¶ÀÁ¢·þÎñºÍ¶ÀÁ¢³Ö¾Ã»¯´æ´¢±á»å´Ú²õÔªÊý¾ÝµÄ¶ÀÁ¢·þÎñºÍ¶ÀÁ¢³Ö¾Ã»¯´æ´¢
±á»å´Ú²õÔªÊý¾ÝµÄ¶ÀÁ¢·þÎñºÍ¶ÀÁ¢³Ö¾Ã»¯´æ´¢
?