[jira] [Commented] (BEAM-7866) Python MongoDB IO performance and correctness issues
[ https://issues.apache.org/jira/browse/BEAM-7866?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16907646#comment-16907646 ] yifan zou commented on BEAM-7866: - The PR got merged. I mark this ticket as resolved. > Python MongoDB IO performance and correctness issues > > > Key: BEAM-7866 > URL: https://issues.apache.org/jira/browse/BEAM-7866 > Project: Beam > Issue Type: Bug > Components: sdk-py-core >Reporter: Eugene Kirpichov >Assignee: Yichi Zhang >Priority: Major > Fix For: 2.15.0 > > Time Spent: 12h > Remaining Estimate: 0h > > https://github.com/apache/beam/blob/master/sdks/python/apache_beam/io/mongodbio.py > splits the query result by computing number of results in constructor, and > then in each reader re-executing the whole query and getting an index > sub-range of those results. > This is broken in several critical ways: > - The order of query results returned by find() is not necessarily > deterministic, so the idea of index ranges on it is meaningless: each shard > may basically get random, possibly overlapping subsets of the total results > - Even if you add order by `_id`, the database may be changing concurrently > to reading and splitting. E.g. if the database contained documents with ids > 10 20 30 40 50, and this was split into shards 0..2 and 3..5 (under the > assumption that these shards would contain respectively 10 20 30, and 40 50), > and then suppose shard 10 20 30 is read and then document 25 is inserted - > then the 3..5 shard will read 30 40 50, i.e. document 30 is duplicated and > document 25 is lost. > - Every shard re-executes the query and skips the first start_offset items, > which in total is quadratic complexity > - The query is first executed in the constructor in order to count results, > which 1) means the constructor can be super slow and 2) it won't work at all > if the database is unavailable at the time the pipeline is constructed (e.g. > if this is a template). > Unfortunately, none of these issues are caught by SourceTestUtils: this class > has extensive coverage with it, and the tests pass. This is because the tests > return the same results in the same order. I don't know how to catch this > automatically, and I don't know how to catch the performance issue > automatically, but these would all be important follow-up items after the > actual fix. > CC: [~chamikara] as reviewer. -- This message was sent by Atlassian JIRA (v7.6.14#76016)
[jira] [Commented] (BEAM-7866) Python MongoDB IO performance and correctness issues
[ https://issues.apache.org/jira/browse/BEAM-7866?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16901247#comment-16901247 ] Yichi Zhang commented on BEAM-7866: --- PR is under review, hopefully I can get this done this week. > Python MongoDB IO performance and correctness issues > > > Key: BEAM-7866 > URL: https://issues.apache.org/jira/browse/BEAM-7866 > Project: Beam > Issue Type: Bug > Components: sdk-py-core >Reporter: Eugene Kirpichov >Assignee: Yichi Zhang >Priority: Blocker > Fix For: 2.15.0 > > Time Spent: 3h 10m > Remaining Estimate: 0h > > https://github.com/apache/beam/blob/master/sdks/python/apache_beam/io/mongodbio.py > splits the query result by computing number of results in constructor, and > then in each reader re-executing the whole query and getting an index > sub-range of those results. > This is broken in several critical ways: > - The order of query results returned by find() is not necessarily > deterministic, so the idea of index ranges on it is meaningless: each shard > may basically get random, possibly overlapping subsets of the total results > - Even if you add order by `_id`, the database may be changing concurrently > to reading and splitting. E.g. if the database contained documents with ids > 10 20 30 40 50, and this was split into shards 0..2 and 3..5 (under the > assumption that these shards would contain respectively 10 20 30, and 40 50), > and then suppose shard 10 20 30 is read and then document 25 is inserted - > then the 3..5 shard will read 30 40 50, i.e. document 30 is duplicated and > document 25 is lost. > - Every shard re-executes the query and skips the first start_offset items, > which in total is quadratic complexity > - The query is first executed in the constructor in order to count results, > which 1) means the constructor can be super slow and 2) it won't work at all > if the database is unavailable at the time the pipeline is constructed (e.g. > if this is a template). > Unfortunately, none of these issues are caught by SourceTestUtils: this class > has extensive coverage with it, and the tests pass. This is because the tests > return the same results in the same order. I don't know how to catch this > automatically, and I don't know how to catch the performance issue > automatically, but these would all be important follow-up items after the > actual fix. > CC: [~chamikara] as reviewer. -- This message was sent by Atlassian JIRA (v7.6.14#76016)
[jira] [Commented] (BEAM-7866) Python MongoDB IO performance and correctness issues
[ https://issues.apache.org/jira/browse/BEAM-7866?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16901243#comment-16901243 ] yifan zou commented on BEAM-7866: - Any ETA on this? [~yichi] > Python MongoDB IO performance and correctness issues > > > Key: BEAM-7866 > URL: https://issues.apache.org/jira/browse/BEAM-7866 > Project: Beam > Issue Type: Bug > Components: sdk-py-core >Reporter: Eugene Kirpichov >Assignee: Yichi Zhang >Priority: Blocker > Fix For: 2.15.0 > > Time Spent: 3h 10m > Remaining Estimate: 0h > > https://github.com/apache/beam/blob/master/sdks/python/apache_beam/io/mongodbio.py > splits the query result by computing number of results in constructor, and > then in each reader re-executing the whole query and getting an index > sub-range of those results. > This is broken in several critical ways: > - The order of query results returned by find() is not necessarily > deterministic, so the idea of index ranges on it is meaningless: each shard > may basically get random, possibly overlapping subsets of the total results > - Even if you add order by `_id`, the database may be changing concurrently > to reading and splitting. E.g. if the database contained documents with ids > 10 20 30 40 50, and this was split into shards 0..2 and 3..5 (under the > assumption that these shards would contain respectively 10 20 30, and 40 50), > and then suppose shard 10 20 30 is read and then document 25 is inserted - > then the 3..5 shard will read 30 40 50, i.e. document 30 is duplicated and > document 25 is lost. > - Every shard re-executes the query and skips the first start_offset items, > which in total is quadratic complexity > - The query is first executed in the constructor in order to count results, > which 1) means the constructor can be super slow and 2) it won't work at all > if the database is unavailable at the time the pipeline is constructed (e.g. > if this is a template). > Unfortunately, none of these issues are caught by SourceTestUtils: this class > has extensive coverage with it, and the tests pass. This is because the tests > return the same results in the same order. I don't know how to catch this > automatically, and I don't know how to catch the performance issue > automatically, but these would all be important follow-up items after the > actual fix. > CC: [~chamikara] as reviewer. -- This message was sent by Atlassian JIRA (v7.6.14#76016)
[jira] [Commented] (BEAM-7866) Python MongoDB IO performance and correctness issues
[ https://issues.apache.org/jira/browse/BEAM-7866?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16898231#comment-16898231 ] Yichi Zhang commented on BEAM-7866: --- I looked into the split key methods using splitVector in java previously, and found no documentation of how to use python client to do the same thing. After many trial and errors I figure out how to call the same api, and will change the split to using splitVector. > Python MongoDB IO performance and correctness issues > > > Key: BEAM-7866 > URL: https://issues.apache.org/jira/browse/BEAM-7866 > Project: Beam > Issue Type: Bug > Components: sdk-py-core >Reporter: Eugene Kirpichov >Assignee: Yichi Zhang >Priority: Blocker > Fix For: 2.14.0, 2.15.0 > > > https://github.com/apache/beam/blob/master/sdks/python/apache_beam/io/mongodbio.py > splits the query result by computing number of results in constructor, and > then in each reader re-executing the whole query and getting an index > sub-range of those results. > This is broken in several critical ways: > - The order of query results returned by find() is not necessarily > deterministic, so the idea of index ranges on it is meaningless: each shard > may basically get random, possibly overlapping subsets of the total results > - Even if you add order by `_id`, the database may be changing concurrently > to reading and splitting. E.g. if the database contained documents with ids > 10 20 30 40 50, and this was split into shards 0..2 and 3..5 (under the > assumption that these shards would contain respectively 10 20 30, and 40 50), > and then suppose shard 10 20 30 is read and then document 25 is inserted - > then the 3..5 shard will read 30 40 50, i.e. document 30 is duplicated and > document 25 is lost. > - Every shard re-executes the query and skips the first start_offset items, > which in total is quadratic complexity > - The query is first executed in the constructor in order to count results, > which 1) means the constructor can be super slow and 2) it won't work at all > if the database is unavailable at the time the pipeline is constructed (e.g. > if this is a template). > Unfortunately, none of these issues are caught by SourceTestUtils: this class > has extensive coverage with it, and the tests pass. This is because the tests > return the same results in the same order. I don't know how to catch this > automatically, and I don't know how to catch the performance issue > automatically, but these would all be important follow-up items after the > actual fix. > CC: [~chamikara] as reviewer. -- This message was sent by Atlassian JIRA (v7.6.14#76016)
[jira] [Commented] (BEAM-7866) Python MongoDB IO performance and correctness issues
[ https://issues.apache.org/jira/browse/BEAM-7866?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16897562#comment-16897562 ] Eugene Kirpichov commented on BEAM-7866: Yes, splitting by range of id is the right thing to do. And the range of ids has to be captured during split() rather than during constructor - constructor can not do any IO. I don't see anything in MongoDB documentation saying that the natural order is deterministic; https://stackoverflow.com/questions/15467099/mongodb-store-and-select-order says that it is not. > Python MongoDB IO performance and correctness issues > > > Key: BEAM-7866 > URL: https://issues.apache.org/jira/browse/BEAM-7866 > Project: Beam > Issue Type: Bug > Components: sdk-py-core >Reporter: Eugene Kirpichov >Assignee: Yichi Zhang >Priority: Blocker > Fix For: 2.14.0 > > > https://github.com/apache/beam/blob/master/sdks/python/apache_beam/io/mongodbio.py > splits the query result by computing number of results in constructor, and > then in each reader re-executing the whole query and getting an index > sub-range of those results. > This is broken in several critical ways: > - The order of query results returned by find() is not necessarily > deterministic, so the idea of index ranges on it is meaningless: each shard > may basically get random, possibly overlapping subsets of the total results > - Even if you add order by `_id`, the database may be changing concurrently > to reading and splitting. E.g. if the database contained documents with ids > 10 20 30 40 50, and this was split into shards 0..2 and 3..5 (under the > assumption that these shards would contain respectively 10 20 30, and 40 50), > and then suppose shard 10 20 30 is read and then document 25 is inserted - > then the 3..5 shard will read 30 40 50, i.e. document 30 is duplicated and > document 25 is lost. > - Every shard re-executes the query and skips the first start_offset items, > which in total is quadratic complexity > - The query is first executed in the constructor in order to count results, > which 1) means the constructor can be super slow and 2) it won't work at all > if the database is unavailable at the time the pipeline is constructed (e.g. > if this is a template). > Unfortunately, none of these issues are caught by SourceTestUtils: this class > has extensive coverage with it, and the tests pass. This is because the tests > return the same results in the same order. I don't know how to catch this > automatically, and I don't know how to catch the performance issue > automatically, but these would all be important follow-up items after the > actual fix. > CC: [~chamikara] as reviewer. -- This message was sent by Atlassian JIRA (v7.6.14#76016)
[jira] [Commented] (BEAM-7866) Python MongoDB IO performance and correctness issues
[ https://issues.apache.org/jira/browse/BEAM-7866?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16897555#comment-16897555 ] Chamikara Jayalath commented on BEAM-7866: -- Thanks Eugene for pointing out these issues. Seems like, in the Java implementation, we build splits with specific IDs. Probably we can do something similar for Python ? [https://github.com/apache/beam/blob/master/sdks/java/io/mongodb/src/main/java/org/apache/beam/sdk/io/mongodb/MongoDbIO.java#L519] BTW is it correct to assume that the source will be correct if that MongoDB instance stays static (without any data insertion/deletion/update) for the duration of the execution of the read step ? (I understand that this might be improbable for many applications of the Database). Seems like elements will have the same order for different find() calls if the database is static since MongoDB defaults to natural sort order. [https://docs.mongodb.com/manual/reference/glossary/#term-natural-order] > Python MongoDB IO performance and correctness issues > > > Key: BEAM-7866 > URL: https://issues.apache.org/jira/browse/BEAM-7866 > Project: Beam > Issue Type: Bug > Components: sdk-py-core >Reporter: Eugene Kirpichov >Assignee: Yichi Zhang >Priority: Blocker > Fix For: 2.14.0 > > > https://github.com/apache/beam/blob/master/sdks/python/apache_beam/io/mongodbio.py > splits the query result by computing number of results in constructor, and > then in each reader re-executing the whole query and getting an index > sub-range of those results. > This is broken in several critical ways: > - The order of query results returned by find() is not necessarily > deterministic, so the idea of index ranges on it is meaningless: each shard > may basically get random, possibly overlapping subsets of the total results > - Even if you add order by `_id`, the database may be changing concurrently > to reading and splitting. E.g. if the database contained documents with ids > 10 20 30 40 50, and this was split into shards 0..2 and 3..5 (under the > assumption that these shards would contain respectively 10 20 30, and 40 50), > and then suppose shard 10 20 30 is read and then document 25 is inserted - > then the 3..5 shard will read 30 40 50, i.e. document 30 is duplicated and > document 25 is lost. > - Every shard re-executes the query and skips the first start_offset items, > which in total is quadratic complexity > - The query is first executed in the constructor in order to count results, > which 1) means the constructor can be super slow and 2) it won't work at all > if the database is unavailable at the time the pipeline is constructed (e.g. > if this is a template). > Unfortunately, none of these issues are caught by SourceTestUtils: this class > has extensive coverage with it, and the tests pass. This is because the tests > return the same results in the same order. I don't know how to catch this > automatically, and I don't know how to catch the performance issue > automatically, but these would all be important follow-up items after the > actual fix. > CC: [~chamikara] as reviewer. -- This message was sent by Atlassian JIRA (v7.6.14#76016)
[jira] [Commented] (BEAM-7866) Python MongoDB IO performance and correctness issues
[ https://issues.apache.org/jira/browse/BEAM-7866?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16897522#comment-16897522 ] Yichi Zhang commented on BEAM-7866: --- Thanks for looking into this, I'll try to address them asap. > Python MongoDB IO performance and correctness issues > > > Key: BEAM-7866 > URL: https://issues.apache.org/jira/browse/BEAM-7866 > Project: Beam > Issue Type: Bug > Components: sdk-py-core >Reporter: Eugene Kirpichov >Assignee: Yichi Zhang >Priority: Blocker > Fix For: 2.14.0 > > > https://github.com/apache/beam/blob/master/sdks/python/apache_beam/io/mongodbio.py > splits the query result by computing number of results in constructor, and > then in each reader re-executing the whole query and getting an index > sub-range of those results. > This is broken in several critical ways: > - The order of query results returned by find() is not necessarily > deterministic, so the idea of index ranges on it is meaningless: each shard > may basically get random, possibly overlapping subsets of the total results > - Even if you add order by `_id`, the database may be changing concurrently > to reading and splitting. E.g. if the database contained documents with ids > 10 20 30 40 50, and this was split into shards 0..2 and 3..5 (under the > assumption that these shards would contain respectively 10 20 30, and 40 50), > and then suppose shard 10 20 30 is read and then document 25 is inserted - > then the 3..5 shard will read 30 40 50, i.e. document 30 is duplicated and > document 25 is lost. > - Every shard re-executes the query and skips the first start_offset items, > which in total is quadratic complexity > - The query is first executed in the constructor in order to count results, > which 1) means the constructor can be super slow and 2) it won't work at all > if the database is unavailable at the time the pipeline is constructed (e.g. > if this is a template). > Unfortunately, none of these issues are caught by SourceTestUtils: this class > has extensive coverage with it, and the tests pass. This is because the tests > return the same results in the same order. I don't know how to catch this > automatically, and I don't know how to catch the performance issue > automatically, but these would all be important follow-up items after the > actual fix. > CC: [~chamikara] as reviewer. -- This message was sent by Atlassian JIRA (v7.6.14#76016)
