[
https://issues.apache.org/jira/browse/BAHIR-213?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16902385#comment-16902385
]
ASF GitHub Bot commented on BAHIR-213:
--------------------------------------
abhishekd0907 commented on pull request #91: [BAHIR-213] Faster S3 file Source
for Structured Streaming with SQS
URL: https://github.com/apache/bahir/pull/91
## What changes were proposed in this pull request?
Using FileStreamSource to read files from a S3 bucket has problems both in
terms of costs and latency:
- **Latency**: Listing all the files in S3 buckets every microbatch can be
both slow and resource intensive.
- **Costs**: Making List API requests to S3 every microbatch can be costly.
The solution is to use Amazon Simple Queue Service (SQS) which lets you find
new files written to S3 bucket without the need to list all the files every
microbatch.
S3 buckets can be configured to send notification to an Amazon SQS Queue on
Object Create / Object Delete events. For details see AWS documentation here
[Configuring S3 Event
Notifications](https://docs.aws.amazon.com/AmazonS3/latest/dev/NotificationHowTo.html)
Spark can leverage this to find new files written to S3 bucket by reading
notifications from SQS queue instead of listing files every microbatch.
This PR adds a new SQSSource which uses Amazon SQS queue to find new files
every microbatch.
## Usage
`val inputDf = spark
.readStream
.format("s3-sqs")
.schema(schema)
.option("fileFormat", "json")
.option("sqsUrl", "https://QUEUE_URL";)
.option("region", "us-east-1")
.load()`
## Implementation Details
We create a scheduled thread which runs asynchronously with the streaming
query thread and periodically fetches messages from the SQS Queue. Key
information related to file path & timestamp is extracted from the SQS messages
and the new files are stored in a thread safe SQS file cache.
Streaming Query thread gets the files from SQS File Cache and filters out
the new files. Based on the maxFilesPerTrigger condition, all or a part of the
new files are added to the offset log and marked as processed in the SQS File
Cache. The corresponding SQS messages for the processed files are deleted from
the Amazon SQS Queue and the offset value is incremented and returned.
## How was this patch tested?
Added new unit tests in SqsSourceOptionsSuite which test various
SqsSourceOptions. Will add more tests after some initial feedback on design
approach and functionality.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
> Faster S3 file Source for Structured Streaming with SQS
> -------------------------------------------------------
>
> Key: BAHIR-213
> URL: https://issues.apache.org/jira/browse/BAHIR-213
> Project: Bahir
> Issue Type: New Feature
> Components: Spark Structured Streaming Connectors
> Affects Versions: Spark-2.4.0
> Reporter: Abhishek Dixit
> Priority: Major
>
> Using FileStreamSource to read files from a S3 bucket has problems both in
> terms of costs and latency:
> * *Latency:* Listing all the files in S3 buckets every microbatch can be
> both slow and resource intensive.
> * *Costs:* Making List API requests to S3 every microbatch can be costly.
> The solution is to use Amazon Simple Queue Service (SQS) which lets you find
> new files written to S3 bucket without the need to list all the files every
> microbatch.
> S3 buckets can be configured to send notification to an Amazon SQS Queue on
> Object Create / Object Delete events. For details see AWS documentation here
> [Configuring S3 Event
> Notifications|https://docs.aws.amazon.com/AmazonS3/latest/dev/NotificationHowTo.html]
>
> Spark can leverage this to find new files written to S3 bucket by reading
> notifications from SQS queue instead of listing files every microbatch.
> I hope to contribute changes proposed in [this pull
> request|https://github.com/apache/spark/pull/24934] to Apache Bahir as
> suggested by [gaborgsomogyi|https://github.com/gaborgsomogyi]
> [here|https://github.com/apache/spark/pull/24934#issuecomment-511389130]
--
This message was sent by Atlassian JIRA
(v7.6.14#76016)
