I have copy and pasted snippets from:
http://hadoop.apache.org/common/docs/current/hdfs_design.html
Note: I added my proprietary, copyrighted, handcrafted and uniquely Vancouver, BC, < fail > < /fail > tags. Note each tag is created with a physical serialized twin linked using the Imaginary Quantum Entanglement DRM. In addition to the < fail > < /fail > tags I also sell premium < FAIL > < /FAIL > tags. Email joshua@neocodesoftware.com to order yours today. (note: don't be fooled by the "counterfeit" < Fail > < /Fail > tags - these are open source, not imaginarily quantumly entangled, and not commercially supported knock offs made by my clone.)
Hadoop - it says highly fault tolerant but does it ACT highly fault tolerant?
1. Introduction
The Hadoop Distributed File System (HDFS) is a distributed file system designed to run on commodity hardware... HDFS is highly fault-tolerant and is designed to be deployed on low-cost hardware.
2. NameNode and DataNodes
HDFS has a master/slave architecture. < fail >An HDFS cluster consists of a single NameNode< /fail >, a master server that manages the file system namespace and regulates access to files by clients.
3. Metadata Disk Failure
< fail >The NameNode machine is a single point of failure for an HDFS cluster.< /fail > If the NameNode machine fails, manual intervention is necessary. Currently, automatic restart and failover of the NameNode software to another machine is not supported.
Conclusion: Not fault tolerant.
Solution: Use DNS server architecture and potentially code and binaries to implement a NameNode that is not a single point of failure, does not need manual intervention.
http://hadoop.apache.org/common/docs/current/hdfs_design.html
Note: I added my proprietary, copyrighted, handcrafted and uniquely Vancouver, BC, < fail > < /fail > tags. Note each tag is created with a physical serialized twin linked using the Imaginary Quantum Entanglement DRM. In addition to the < fail > < /fail > tags I also sell premium < FAIL > < /FAIL > tags. Email joshua@neocodesoftware.com to order yours today. (note: don't be fooled by the "counterfeit" < Fail > < /Fail > tags - these are open source, not imaginarily quantumly entangled, and not commercially supported knock offs made by my clone.)
Hadoop - it says highly fault tolerant but does it ACT highly fault tolerant?
1. Introduction
The Hadoop Distributed File System (HDFS) is a distributed file system designed to run on commodity hardware... HDFS is highly fault-tolerant and is designed to be deployed on low-cost hardware.
2. NameNode and DataNodes
HDFS has a master/slave architecture. < fail >An HDFS cluster consists of a single NameNode< /fail >, a master server that manages the file system namespace and regulates access to files by clients.
3. Metadata Disk Failure
< fail >The NameNode machine is a single point of failure for an HDFS cluster.< /fail > If the NameNode machine fails, manual intervention is necessary. Currently, automatic restart and failover of the NameNode software to another machine is not supported.
Conclusion: Not fault tolerant.
Solution: Use DNS server architecture and potentially code and binaries to implement a NameNode that is not a single point of failure, does not need manual intervention.
Comments