[ https://issues.apache.org/jira/browse/HADOOP-16132?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16780886#comment-16780886 ]
Justin Uang commented on HADOOP-16132: -------------------------------------- Copying over the last comment from the github ticket since we will be continuing the conversation here: [~ste...@apache.org] {quote}BTW, one little side effect of breaking up the reads: every GET is its own HTTP request, so gets billed differently, and for SSE-KMS, possibly a separate call to AWS:KMS. Nobody quite knows about the latter, we do know that heavy random seek IO on a single tree in a bucket can trigger more throttling than you'd expect Anyway, maybe for random IO the strategy would be to have a notion of aligned blocks, say 8 MB, the current block is cached as it is read in, so a backward seek can often work from in memory; the stream could be doing a readahead of , say, the next 2+ blocks in parallel & then store them in a ring of cached blocks ready for when they are used. you've got me thinking now... {quote} > Support multipart download in S3AFileSystem > ------------------------------------------- > > Key: HADOOP-16132 > URL: https://issues.apache.org/jira/browse/HADOOP-16132 > Project: Hadoop Common > Issue Type: Sub-task > Components: fs/s3 > Reporter: Justin Uang > Priority: Major > Attachments: HADOOP-16132.001.patch > > > I noticed that I get 150MB/s when I use the AWS CLI > {code:java} > aws s3 cp s3://<bucket>/<key> - > /dev/null{code} > vs 50MB/s when I use the S3AFileSystem > {code:java} > hadoop fs -cat s3://<bucket>/<key> > /dev/null{code} > Looking into the AWS CLI code, it looks like the > [download|https://github.com/boto/s3transfer/blob/ca0b708ea8a6a1213c6e21ca5a856e184f824334/s3transfer/download.py] > logic is quite clever. It downloads the next couple parts in parallel using > range requests, and then buffers them in memory in order to reorder them and > expose a single contiguous stream. I translated the logic to Java and > modified the S3AFileSystem to do similar things, and am able to achieve > 150MB/s download speeds as well. It is mostly done but I have some things to > clean up first. The PR is here: > https://github.com/palantir/hadoop/pull/47/files > It would be great to get some other eyes on it to see what we need to do to > get it merged. -- This message was sent by Atlassian JIRA (v7.6.3#76005) --------------------------------------------------------------------- To unsubscribe, e-mail: common-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: common-issues-h...@hadoop.apache.org