[
https://issues.apache.org/jira/browse/HADOOP-11183?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Thomas Demoor updated HADOOP-11183:
-----------------------------------
Attachment: info-S3AFastOutputStream-sync.md
HADOOP-11183.001.patch
Patch 001: synchronous implementation (blocks on every partUpload). A little
extra info is provided in info-S3AFastOutputStream-sync.md.
Additional remarks:
# This patch is simply to set the stage and kick off the discussion. I am
working on an async version (multiple concurrent partuploads), which I will
post asap.
# I would really like to bump up the aws-sdk version but in some other jira
this was said to be a gargantuan task (probably http versions conflicting with
other libraries? azure?) Correct?
# I also renamed partSizeThreshold to the correct term multiPartThreshold.
(creating a separate issue seemed overkill).
> Memory-based S3AOutputstream
> ----------------------------
>
> Key: HADOOP-11183
> URL: https://issues.apache.org/jira/browse/HADOOP-11183
> Project: Hadoop Common
> Issue Type: Improvement
> Components: fs/s3
> Affects Versions: 2.6.0
> Reporter: Thomas Demoor
> Attachments: HADOOP-11183.001.patch, info-S3AFastOutputStream-sync.md
>
>
> Currently s3a buffers files on disk(s) before uploading. This JIRA
> investigates adding a memory-based upload implementation.
> The motivation is evidently performance: this would be beneficial for users
> with high network bandwidth to S3 (EC2?) or users that run Hadoop directly on
> an S3-compatible object store (FYI: my contributions are made in name of
> Amplidata).
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
