You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@nifi.apache.org by "Mr TheSegfault (JIRA)" <ji...@apache.org> on 2019/03/18 11:27:00 UTC
[jira] [Commented] (MINIFICPP-769) In-memory content storage for
Nanofi
[ https://issues.apache.org/jira/browse/MINIFICPP-769?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16794944#comment-16794944 ]
Mr TheSegfault commented on MINIFICPP-769:
------------------------------------------
For NanoFi a volatile repo that avoids as much synchronization as possible and doesn't write to local disk ( when it makes sense ) is a great idea. Given that this will live NanoFi and be written in C do you have a feeling for if this should be thread safe at all? It's okay if it is explicitly not, just curious abut the plan.
> In-memory content storage for Nanofi
> ------------------------------------
>
> Key: MINIFICPP-769
> URL: https://issues.apache.org/jira/browse/MINIFICPP-769
> Project: NiFi MiNiFi C++
> Issue Type: Brainstorming
> Reporter: Arpad Boda
> Assignee: Arpad Boda
> Priority: Major
> Labels: nanofi
>
> Currently Nanofi supports storing flow file content in:
> -Files (content location is a file system path)
> -C++ content repo
> The latter is planned to go away as we move towards C, however an IO-friendly way of processing and forwarding data would be nice.
> For eg. in case the data source is bluetooth/network/pipe, we should be able to provide a way of passing it through some processors and sending it via S2S without writing the content to local storage.
> [~phrocker] , [~aldrin] what's your opinion?
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)