hama-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Hyunsik Choi (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HAMA-201) Design the storage structure for graph data
Date Mon, 19 Oct 2009 09:04:31 GMT

    [ https://issues.apache.org/jira/browse/HAMA-201?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12767249#action_12767249

Hyunsik Choi commented on HAMA-201:

In order to complete the skeleton structure of the angrapa, first of all we need at least
dummy components that constitute the angrapa. Among them, the storage structure is one of
the most necessary components.

In terms of this point of view, It would be good that Initial storage structure is based on
Hbase. Although Hbase will be not optimal to angrapa, Hbase satisfies most of the requirements,
and the storage structure based on Hbase will be implemented relatively easier than others.

What do you think about that?

> Design the storage structure for graph data
> -------------------------------------------
>                 Key: HAMA-201
>                 URL: https://issues.apache.org/jira/browse/HAMA-201
>             Project: Hama
>          Issue Type: New Feature
>          Components: graph
>    Affects Versions: 0.2.0
>            Reporter: Hyunsik Choi
>            Assignee: Hyunsik Choi
>             Fix For: 0.2.0
> For graph package, we need a novel storage structure in response to our main objective.
It can be a file structure or a storage system like Hbase.
> Its requirements are as follows:
>  * It has to support random access to any vertex by a given key.
>  * The key will be an array of bytes.
>  * A vertex or edge has a label value composed of an array of bytes.
>  * The maximum degree of a vertex can be almost infinity.

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

View raw message