What is the biggest problem of extremely large hadoop cluster ?

Hi , I am curious to know what is the biggest problem of extremely large hadoop cluster. What I can imagine now is the memory cost of meta data of hdfs in name node. One solution I can think about now is to use other storage implementation such as database to store the metadata, although it has pe

[jira] Created: (HDFS-994) Provide methods for obtaining delegation token from Namenode for hftp and other uses

Provide methods for obtaining delegation token from Namenode for hftp and other uses Key: HDFS-994 URL: https://issues.apache.org/jira/browse/HDFS-994 Project: Hadoo

[jira] Created: (HDFS-995) Replace usage of FileStatus#isDir()

Replace usage of FileStatus#isDir() --- Key: HDFS-995 URL: https://issues.apache.org/jira/browse/HDFS-995 Project: Hadoop HDFS Issue Type: Improvement Components: name-node Affects Versions: 0.22