[ https://issues.apache.org/jira/browse/HDFS-599?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12753240#action_12753240 ]
Hong Tang commented on HDFS-599: -------------------------------- Using different ports for QoS also has its limitations. The request dispatching logic must be baked in the client side. Inside the server, multiple threads or multiple thread pools that handle different classes are oblivious to each other and may not enforce fine grain resource control.This OSDI paper (which I co-authored), http://www.usenix.org/events/osdi02/tech/shen.html, describes a complicated case of QoS policy and the solution is based on queues. Switches may be useful for network QoS (e.g. preventing DoS attacks and prioritize traffic). But for application level QoS (where the system is not constrained by the amount of network traffic, but the work needs to be performed in the server), then I'd think queue-based solution is better. > Improve Namenode robustness by prioritizing datanode heartbeats over client > requests > ------------------------------------------------------------------------------------ > > Key: HDFS-599 > URL: https://issues.apache.org/jira/browse/HDFS-599 > Project: Hadoop HDFS > Issue Type: Improvement > Components: name-node > Reporter: dhruba borthakur > Assignee: dhruba borthakur > > The namenode processes RPC requests from clients that are reading/writing to > files as well as heartbeats/block reports from datanodes. > Sometime, because of various reasons (Java GC runs, inconsistent performance > of NFS filer that stores HDFS transacttion logs, etc), the namenode > encounters transient slowness. For example, if the device that stores the > HDFS transaction logs becomes sluggish, the Namenode's ability to process > RPCs slows down to a certain extent. During this time, the RPCs from clients > as well as the RPCs from datanodes suffer in similar fashion. If the > underlying problem becomes worse, the NN's ability to process a heartbeat > from a DN is severly impacted, thus causing the NN to declare that the DN is > dead. Then the NN starts replicating blocks that used to reside on the > now-declared-dead datanode. This adds extra load to the NN. Then the > now-declared-datanode finally re-establishes contact with the NN, and sends a > block report. The block report processing on the NN is another heavyweight > activity, thus casing more load to the already overloaded namenode. > My proposal is tha the NN should try its best to continue processing RPCs > from datanodes and give lesser priority to serving client requests. The > Datanode RPCs are integral to the consistency and performance of the Hadoop > file system, and it is better to protect it at all costs. This will ensure > that NN recovers from the hiccup much faster than what it does now. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.