hadoop - File I/O on NoSQL - especially HBase - is it recommended? or not? -
i'm new @ nosql , i'm trying use hbase file storage. i'll store files in hbase binary.
i don't need statistics, file storage.
is recommended? worry i/o speed.
the reason why use hbase storage have use hdfs, can't build hadoop on client computer. because of it, tring find libraries helps client connect hdfs files. couldn't find it, , choose hbase instead of connection library.
in situation, should do?
i don't know hadoop, mongodb has gridfs designed distributed file storage enables scale horizontally, replication "free" , on.
http://www.mongodb.org/display/docs/gridfs
there overhead storing files in chunks in mongodb, if load low medium, , need low response times, better off using file system directly. performance vary between different driver implementations.
Comments
Post a Comment