[jira] [Commented] (HADOOP-15489) S3Guard to self update on directory listings of S3
[ https://issues.apache.org/jira/browse/HADOOP-15489?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16622914#comment-16622914 ] Aaron Fabbri commented on HADOOP-15489: --- [~ste...@apache.org] that was the original behavior and we changed it to improve performance (those constant write-backs of listings into dynamo become expensive). I think it is fairly close to reasonable default behavior now, but I'd be ok with adding a non-default config knob to get the "always write back listings" behavior. I'm thinking it might be slow though. All depends on the access patterns. > S3Guard to self update on directory listings of S3 > -- > > Key: HADOOP-15489 > URL: https://issues.apache.org/jira/browse/HADOOP-15489 > Project: Hadoop Common > Issue Type: Sub-task > Components: fs/s3 >Affects Versions: 3.1.0 > Environment: s3guard >Reporter: Steve Loughran >Assignee: Gabor Bota >Priority: Major > > S3Guard updates its table on a getFileStatus call, but not on a directory > listing. > While this makes directory listings faster (no need to push out an update), > it slows down subsequent queries of the files, such as a sequence of: > {code} > statuses = s3a.listFiles(dir) > for (status: statuses) { > if (status.isFile) { > try(is = s3a.open(status.getPath())) { > ... do something > } > } > {code} > this is because the open() is doing the getFileStatus check, even after the > listing. > Updating the DDB tables after a listing would give those reads a speedup, > albeit at the expense of initiating a (bulk) update in the list call. Of > course, we could consider making that async, though that design (essentially > a write-buffer) would require the buffer to be checked in the reads too. -- 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
[jira] [Commented] (HADOOP-15489) S3Guard to self update on directory listings of S3
[ https://issues.apache.org/jira/browse/HADOOP-15489?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16549672#comment-16549672 ] Steve Loughran commented on HADOOP-15489: - OK, I see, if !auth no update. Explains why I didn't see it. I think I'd like this even in non-auth, probably, so that if you do a getFileStatus() after it can bypass the FS. But: i see the risks here if something else does an update > S3Guard to self update on directory listings of S3 > -- > > Key: HADOOP-15489 > URL: https://issues.apache.org/jira/browse/HADOOP-15489 > Project: Hadoop Common > Issue Type: Sub-task > Components: fs/s3 >Affects Versions: 3.1.0 > Environment: s3guard >Reporter: Steve Loughran >Priority: Major > > S3Guard updates its table on a getFileStatus call, but not on a directory > listing. > While this makes directory listings faster (no need to push out an update), > it slows down subsequent queries of the files, such as a sequence of: > {code} > statuses = s3a.listFiles(dir) > for (status: statuses) { > if (status.isFile) { > try(is = s3a.open(status.getPath())) { > ... do something > } > } > {code} > this is because the open() is doing the getFileStatus check, even after the > listing. > Updating the DDB tables after a listing would give those reads a speedup, > albeit at the expense of initiating a (bulk) update in the list call. Of > course, we could consider making that async, though that design (essentially > a write-buffer) would require the buffer to be checked in the reads too. -- 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
[jira] [Commented] (HADOOP-15489) S3Guard to self update on directory listings of S3
[ https://issues.apache.org/jira/browse/HADOOP-15489?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16548655#comment-16548655 ] Aaron Fabbri commented on HADOOP-15489: --- [~ste...@apache.org] see my comment above. Is there anything to do here? > S3Guard to self update on directory listings of S3 > -- > > Key: HADOOP-15489 > URL: https://issues.apache.org/jira/browse/HADOOP-15489 > Project: Hadoop Common > Issue Type: Sub-task > Components: fs/s3 >Affects Versions: 3.1.0 > Environment: s3guard >Reporter: Steve Loughran >Priority: Major > > S3Guard updates its table on a getFileStatus call, but not on a directory > listing. > While this makes directory listings faster (no need to push out an update), > it slows down subsequent queries of the files, such as a sequence of: > {code} > statuses = s3a.listFiles(dir) > for (status: statuses) { > if (status.isFile) { > try(is = s3a.open(status.getPath())) { > ... do something > } > } > {code} > this is because the open() is doing the getFileStatus check, even after the > listing. > Updating the DDB tables after a listing would give those reads a speedup, > albeit at the expense of initiating a (bulk) update in the list call. Of > course, we could consider making that async, though that design (essentially > a write-buffer) would require the buffer to be checked in the reads too. -- 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
[jira] [Commented] (HADOOP-15489) S3Guard to self update on directory listings of S3
[ https://issues.apache.org/jira/browse/HADOOP-15489?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16490281#comment-16490281 ] Aaron Fabbri commented on HADOOP-15489: --- Look at S3Guard#dirListingUnion(). We used to always update MetadataStore at the end of listStatus(). Later we changed it to only happen when fs.s3a.metadatastore.authoritative = true. If you set this to true you will always update MetadataStore at end of listStatus(), but keep in mind that the short-circuit listings are not implemented for Dynamo MS yet ([~gabor.bota] is working towards that though). There are other listing APIs that don't do this, of course. > S3Guard to self update on directory listings of S3 > -- > > Key: HADOOP-15489 > URL: https://issues.apache.org/jira/browse/HADOOP-15489 > Project: Hadoop Common > Issue Type: Sub-task > Components: fs/s3 >Affects Versions: 3.1.0 > Environment: s3guard >Reporter: Steve Loughran >Priority: Major > > S3Guard updates its table on a getFileStatus call, but not on a directory > listing. > While this makes directory listings faster (no need to push out an update), > it slows down subsequent queries of the files, such as a sequence of: > {code} > statuses = s3a.listFiles(dir) > for (status: statuses) { > if (status.isFile) { > try(is = s3a.open(status.getPath())) { > ... do something > } > } > {code} > this is because the open() is doing the getFileStatus check, even after the > listing. > Updating the DDB tables after a listing would give those reads a speedup, > albeit at the expense of initiating a (bulk) update in the list call. Of > course, we could consider making that async, though that design (essentially > a write-buffer) would require the buffer to be checked in the reads too. -- 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
[jira] [Commented] (HADOOP-15489) S3Guard to self update on directory listings of S3
[ https://issues.apache.org/jira/browse/HADOOP-15489?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16484119#comment-16484119 ] Steve Loughran commented on HADOOP-15489: - No benchmark data here; just realised the issue exists if you are doing things like treewalks of a filesystem not yet imported into S3Guard. Although the S3 LIST calls get the data of the descendants, the DDB table isn't updated. This is underperformant > S3Guard to self update on directory listings of S3 > -- > > Key: HADOOP-15489 > URL: https://issues.apache.org/jira/browse/HADOOP-15489 > Project: Hadoop Common > Issue Type: Sub-task > Components: fs/s3 >Affects Versions: 3.1.0 > Environment: s3guard >Reporter: Steve Loughran >Priority: Major > > S3Guard updates its table on a getFileStatus call, but not on a directory > listing. > While this makes directory listings faster (no need to push out an update), > it slows down subsequent queries of the files, such as a sequence of: > {code} > statuses = s3a.listFiles(dir) > for (status: statuses) { > if (status.isFile) { > try(is = s3a.open(status.getPath())) { > ... do something > } > } > {code} > this is because the open() is doing the getFileStatus check, even after the > listing. > Updating the DDB tables after a listing would give those reads a speedup, > albeit at the expense of initiating a (bulk) update in the list call. Of > course, we could consider making that async, though that design (essentially > a write-buffer) would require the buffer to be checked in the reads too. -- 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