Products & Services
MapR Book Club
to create and rate content, and to follow, bookmark, and share content with other members.
need sequence of steps during data write operation to mapr fs
Question asked by
on Jan 20, 2015
on Jan 21, 2015 by sholdship
Show 0 Likes
I am looking for steps and invloved during data write operation to mapr fs (like how client contact to CLDB and how CLDB manage data write request )
No one else has this question
Mark as assumed answered
This content has been marked as final.
Show 1 comment
(Required, will not be published)
Jan 21, 2015 9:06 AM
CLDB information is aggressively cached by file clients
All IO within MapR file system starts with an Open. For example, if you are looking for a file at the location: /mapr///. The following happens:
The File Client resolves the cluster name > CLDB mapping for by checking the local
mapr-clusters.conf file. The mapr-clusters.conf file provides both cluster name and CLDB servers:ports
The File Client begins querying the CLDB nodes to lookup the name container locations for the
mapr.cluster.root i.e. Looking up / in . The File Client connects to the MFS nodes provided
Once the File Client can connect to one of the MFS Nodes, it looks up in the name container for
mapr.cluster.root. If is indeed a volume link, the FileClient returns to CLDB to get name container locations for the volume.
Looking up in â€œ/", to get FID for â€œ//â€ with the FID of the file, the File Client retrieves metadata for the file (e.g. chunk locations).
For a read, the File Client will read back from any of the MFS nodes containing the chunk of data it needs.
For a write the File Client requests a FID to be allocated. A request is sent to the MFS node with the master copy of the name container for the volume. With the new FID, the File Client sends the write to the
master container, data is replicated to the copies and the acknowledgement is returned to file client.
Show 0 Likes
Retrieving data ...
Using Apache Spark DataFrames for Processing of Tabular Data - Let's Discuss
How to deploy MapR, Mesos, Marathon, Docker and Spark and run your first containers and jobs
Writing Dataframe as MapRDB Binary Table
ERROR Cidcache fs/client/fileclient/cc/cidcache.cc:1047
Kafka Connect vs StreamSets: advantages and disadvantages?