cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jun Rao (JIRA)" <>
Subject [jira] Commented: (CASSANDRA-193) Proactive repair
Date Thu, 12 Nov 2009 01:18:39 GMT


Jun Rao commented on CASSANDRA-193:

2.1 Still confused. So node.depth doesn't mean the depth of the node in the tree?

3.1 The ranges don't change after the iterator is generated, right? But inside the while loop
in Validator.add, there is comment about adding a new range. So, are you really adding a new
range there or the comment is wrong?

3.2 Why is Merkle tree a binary tree? I thought it's an n-ary tree since you can specify the
order of each node.

Since a leaf node corresponds to a single range, why do you need a list of hashes? I thought
each leaf node maintains a single hash value for rows its range and each inner node maintains
a single hash that is the logical AND of the hashes of all its children. Is this not correct?

> Proactive repair
> ----------------
>                 Key: CASSANDRA-193
>                 URL:
>             Project: Cassandra
>          Issue Type: New Feature
>          Components: Core
>            Reporter: Jonathan Ellis
>            Assignee: Stu Hood
>             Fix For: 0.5
>         Attachments: 193-1-tree-preparation.diff, 193-2-tree.diff, 193-3-aes-preparation.diff,
> Currently cassandra supports "read repair," i.e., lazy repair when a read is done.  This
is better than nothing but is not sufficient for some cases (e.g. catastrophic node failure
where you need to rebuild all of a node's data on a new machine).
> Dynamo uses merkle trees here.  This is harder for Cassandra given the CF data model
but I suppose we could just hash the serialized CF value.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message