[jira] [Commented] (SOLR-15051) Shared storage -- BlobDirectory (de-duping)

["Kevin Risden (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SOLR-15051?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17258466#comment-17258466 ] 

Kevin Risden commented on SOLR-15051:
-------------------------------------

{quote}"Hadoop filesystem interface" is an ideal choice{quote}

Not sure I would go so far as an "ideal choice" - it definitely brings in a lot of dependencies sadly. The nice part is that it has implementations for local, hdfs, s3, adls, and gcp at least. So you get multi cloud for free. What I do not know is how efficient each of those implementations are - especially the "local" one for file:// type paths.

Just wanted to point out the Hadoop filesystem connector stuff is separate from the full running of HDFS. Here are a few references:

* General overview: http://hadoop.apache.org/docs/current3/hadoop-project-dist/hadoop-common/filesystem/index.html
* AWS: http://hadoop.apache.org/docs/current3/hadoop-aws/tools/hadoop-aws/index.html
* Azure: http://hadoop.apache.org/docs/current3/hadoop-azure/index.html
* Google: https://github.com/GoogleCloudDataproc/hadoop-connectors
* 

> Shared storage -- BlobDirectory (de-duping)
> -------------------------------------------
>
>                 Key: SOLR-15051
>                 URL: https://issues.apache.org/jira/browse/SOLR-15051
>             Project: Solr
>          Issue Type: Improvement
>      Security Level: Public(Default Security Level. Issues are Public) 
>            Reporter: David Smiley
>            Assignee: David Smiley
>            Priority: Major
>          Time Spent: 10m
>  Remaining Estimate: 0h
>
> This proposal is a way to accomplish shared storage in SolrCloud with a few key characteristics: (A) using a Directory implementation, (B) delegates to a backing local file Directory as a kind of read/write cache (C) replicas have their own "space", (D) , de-duplication across replicas via reference counting, (E) uses ZK but separately from SolrCloud stuff.
> The Directory abstraction is a good one, and helps isolate shared storage from the rest of SolrCloud that doesn't care.  Using a backing normal file Directory is faster for reads and is simpler than Solr's HDFSDirectory's BlockCache.  Replicas having their own space solves the problem of multiple writers (e.g. of the same shard) trying to own and write to the same space, and it implies that any of Solr's replica types can be used along with what goes along with them like peer-to-peer replication (sometimes faster/cheaper than pulling from shared storage).  A de-duplication feature solves needless duplication of files across replicas and from parent shards (i.e. from shard splitting).  The de-duplication feature requires a place to cache directory listings so that they can be shared across replicas and atomically updated; this is handled via ZooKeeper.  Finally, some sort of Solr daemon / auto-scaling code should be added to implement "autoAddReplicas", especially to provide for a scenario where the leader is gone and can't be replicated from directly but we can access shared storage.
> For more about shared storage concepts, consider looking at the description in SOLR-13101 and the linked Google Doc.
> *[PROPOSAL DOC|https://docs.google.com/document/d/1kjQPK80sLiZJyRjek_Edhokfc5q9S3ISvFRM2_YeL8M/edit?usp=sharing]*



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org