On Wed, Jul 17, 2019 at 9:12 PM Gleb Kanterov <[email protected]> wrote: >> >> Suppose one assigns a sharding function to a PCollection. Is it lazy, >> or does it induce a reshuffle right at that point? In either case, >> once the ShardingFn has been applied, how long does it remain in >> effect? Does it prohibit the runner (or user) from doing subsequent >> resharding (including dynamic load balancing)? What happens when one >> has a DoFn that changes the value? (Including the DoFns in our sinks >> that assign random keys.) > > > What if we would reason about sharding in the same way as we reason about > timestamps? > > Please correct me if I am wrong, as I know, in Beam, timestamps exist for > each element. You can get timestamp by using Reify.timestamps. If there are > timestamped values, and they go through ParDo, timestamps are preserved.
That is correct. > We can think of the same with sharding, where Reify.shards would be > PTransform<PCollection<T>, ShardedValue<T>> and ShardedValue<?> would contain > shard and a grouping key. Meaning the shard that the PCollection is currently sharded by, or the one that it should be sharded by in the future. (Your use case is a bit strange in that a single key may be spread across multiple shards, as long as they're part of the same "bucket.") > ParDo wouldn't change sharding and would propagate ShardingFn. The ShardingFn may not be applicable to downstream (mutated) elements. FYI, internally this is handled by having annotations on DoFns as being key-preserving, and only reasoning about operations separated by such DoFns. > CoGroupByKey on such PTransforms would reify grouping key, and do regular > CoGroupByKey, or be rewritten to a regular ParDo if sharding of inputs is > compatible. > > As you mentioned, it requires dynamic work rebalancing to preserve sharding. > What if we do dynamic work rebalancing for each shard independently, as, I > guess, it's done today for fixed windows. Currently, the unit of colocation is by key. Generally sharding introduces a notion of colocation where multiple keys (or mulitple elements, I suppose it need not be keyed) are promised to be processed by the same machine. This is both to constraining (wrt dynamic reshrading) and not needed (with respect to SMB, as your "colocation" is per bucket, but buckets themselves can be processed in a distributed manner). > When we do a split, we would split one shard into two. It should be possible > to do consistently if values within buckets are sorted, in this case, we > would split ranges of possible values. I'm not quite following here. Suppose one processes element a, m, and z. Then one decides to split the bundle, but there's not a "range" we can pick for the "other" as this bundle already spans the whole range. But maybe I'm just off in the weeds here. > On Wed, Jul 17, 2019 at 6:37 PM Robert Bradshaw <[email protected]> wrote: >> >> On Wed, Jul 17, 2019 at 4:26 PM Gleb Kanterov <[email protected]> wrote: >> > >> > I find there is an interesting point in the comments brought by Ahmed >> > Eleryan. Similar to WindowFn, having a concept of ShardingFn, that enables >> > users to implement a class for sharding data. Each Beam node can have >> > ShardingFn set, similar to WindowFn (or WindowingStrategy). Sinks and >> > sources are aware of that and preserve this information. Using that it's >> > possible to do optimization on Beam graph, removing redundant >> > CoGroupByKey, and it would be transparent to users. >> > >> > It feels like a nice addition to the Beam model, or possibly we can >> > implement it using existing windowing mechanics. There are people on the >> > list with strong experience in the area, I'm wondering what do you think? >> >> I've actually thought about this some, though it's been quite a while. >> At the time it seemed hard to work it into a cohesive part of the >> model (even ignoring the fact that sharding is primarily an execution, >> rather than logical, property). >> >> Suppose one assigns a sharding function to a PCollection. Is it lazy, >> or does it induce a reshuffle right at that point? In either case, >> once the ShardingFn has been applied, how long does it remain in >> effect? Does it prohibit the runner (or user) from doing subsequent >> resharding (including dynamic load balancing)? What happens when one >> has a DoFn that changes the value? (Including the DoFns in our sinks >> that assign random keys.) >> >> Right now one can get most of the semantics of sharding by keying by >> the shard id and doing a GBK, where the resulting value set (which is >> allowed to be arbitrarily big) is the (indivisible) shard (e.g. for >> writing to a single file.) >> >> I think sharding (like ordering, the two are quite related) is a >> property that a PCollection can have, and could be leveraged by the >> optimizer, but it's difficult to see how it's propagated through >> transforms. The most sane way to reason about it IMHO is similar to >> sink triggers, where one specifies that one wants a sharding at some >> point, and the runner arranges things upstream such that it is so, and >> some operations can declare that they happen to produce data sharded >> in some way (though again, PCollection to PCollection one needs a >> consistent notion of key to have a consistent notion of sharding). >> >> > Gleb >> > >> > On Tue, Jul 16, 2019 at 11:34 PM Eugene Kirpichov <[email protected]> >> > wrote: >> >> >> >> I'd like to reiterate the request to not build anything on top of >> >> FileBasedSource/Reader. >> >> If the design requires having some interface for representing a function >> >> from a filename to a stream of records, better introduce a new interface >> >> for that. >> >> If it requires interoperability with other IOs that read files, better >> >> change them to use the new interface. >> >> >> >> On Tue, Jul 16, 2019 at 9:08 AM Chamikara Jayalath <[email protected]> >> >> wrote: >> >>> >> >>> Thanks this clarifies a lot. >> >>> >> >>> For writer, I think it's great if you can utilize existing FileIO.Sink >> >>> implementations even if you have to reimplement some of the logic (for >> >>> example compression, temp file handling) that is already implemented in >> >>> Beam FileIO/WriteFiles transforms in your SMB sink transform. >> >>> >> >>> For reader, you are right that there's no FileIO.Read. What we have are >> >>> various implementations of FileBasedSource/FileBasedReader classes that >> >>> are currently intentionally hidden since Beam IO transforms are expected >> >>> to be the intended public interface for users. If you can expose and >> >>> re-use these classes with slight modifications (keeping backwards >> >>> compatibility) I'm OK with it. Otherwise you'll have to write your own >> >>> reader implementations. >> >>> >> >>> In general, seems like SMB has very strong requirements related to >> >>> sharding/hot-key management that are not easily achievable by >> >>> implementing SMB source/sink as a composite transform that utilizes >> >>> existing source/sink transforms. This forces you to implement this logic >> >>> in your own DoFns and existing Beam primitives are not easily re-usable >> >>> in this context. >> >>> >> >>> Thanks, >> >>> Cham >> >>> >> >>> On Tue, Jul 16, 2019 at 8:26 AM Neville Li <[email protected]> wrote: >> >>>> >> >>>> A little clarification of the IO requirement and my understanding of >> >>>> the current state of IO. >> >>>> >> >>>> tl;dr: not sure if there're reusable bits for the reader. It's possible >> >>>> to reuse some for the writer but with heavy refactoring. >> >>>> >> >>>> Reader >> >>>> >> >>>> For each bucket (containing the same key partition, sorted) across >> >>>> multiple input data sets, we stream records from bucket files and merge >> >>>> sort. >> >>>> We open the files in a DoFn, and emit KV<K, CoGbkResult> where the >> >>>> CGBKR encapsulates Iterable<V> from each input. >> >>>> Basically we need a simple API like ResourceId -> Iterator<T>, i.e. >> >>>> sequential read, no block/offset/split requirement. >> >>>> FileBasedSource.FileBasedReader seems the closest fit but they're >> >>>> nested & decoupled. >> >>>> There's no FileIO.Read, only a ReadMatches[1], which can be used with >> >>>> ReadAllViaFileBasedSource<T>. But that's not the granularity we need, >> >>>> since we lose ordering of the input records, and can't merge 2+ sources. >> >>>> >> >>>> Writer >> >>>> >> >>>> We get a `PCollection<BucketShardId, Iterable<T>>` after bucket and and >> >>>> sort, where Iterable<T> is the records sorted by key and BucketShardId >> >>>> is used to produce filename, e.g. bucket-00001-shard-00002.avro. >> >>>> We write each Iterable<T> to a temp file and move to final destination >> >>>> when done. Both should ideally reuse existing code. >> >>>> Looks like FileIO.Sink (and impls in AvroIO, TextIO, TFRecordIO) >> >>>> supports record writing into a WritableByteChannel, but some logic like >> >>>> compression is handled in FileIO through ViaFileBasedSink which extends >> >>>> FileBasedSink. >> >>>> FileIO uses WriteFiles[3] to shard and write of PCollection<T>. Again >> >>>> we lose ordering of the output records or custom file naming scheme. >> >>>> However, WriteShardsIntoTempFilesFn[4] and FinalizeTempFileBundles[5] >> >>>> in WriteFiles seem closest to our need but would have to be split out >> >>>> and generalized. >> >>>> >> >>>> Note on reader block/offset/split requirement >> >>>> >> >>>> Because of the merge sort, we can't split or offset seek a bucket file. >> >>>> Because without persisting the offset index of a key group somewhere, >> >>>> we can't efficiently skip to a key group without exhausting the >> >>>> previous ones. Furthermore we need to merge sort and align keys from >> >>>> multiple sources, which may not have the same key distribution. It >> >>>> might be possible to binary search for matching keys but that's extra >> >>>> complication. IMO the reader work distribution is better solved by >> >>>> better bucket/shard strategy in upstream writer. >> >>>> >> >>>> References >> >>>> >> >>>> ReadMatches extends PTransform<PCollection<MatchResult.Metadata>, >> >>>> PCollection<ReadableFile>> >> >>>> ReadAllViaFileBasedSource<T> extends >> >>>> PTransform<PCollection<ReadableFile>, PCollection<T>> >> >>>> WriteFiles<UserT, DestinationT, OutputT> extends >> >>>> PTransform<PCollection<UserT>, WriteFilesResult<DestinationT>> >> >>>> WriteShardsIntoTempFilesFn extends DoFn<KV<ShardedKey<Integer>, >> >>>> Iterable<UserT>>, FileResult<DestinationT>> >> >>>> FinalizeTempFileBundles extends PTransform< >> >>>> PCollection<List<FileResult<DestinationT>>>, >> >>>> WriteFilesResult<DestinationT>> >> >>>> >> >>>> >> >>>> On Tue, Jul 16, 2019 at 5:15 AM Robert Bradshaw <[email protected]> >> >>>> wrote: >> >>>>> >> >>>>> On Mon, Jul 15, 2019 at 7:03 PM Eugene Kirpichov >> >>>>> <[email protected]> wrote: >> >>>>> > >> >>>>> > Quick note: I didn't look through the document, but please do not >> >>>>> > build on either FileBasedSink or FileBasedReader. They are both >> >>>>> > remnants of the old, non-composable IO world; and in fact much of >> >>>>> > the composable IO work emerged from frustration with their >> >>>>> > limitations and recognizing that many other IOs were suffering from >> >>>>> > the same limitations. >> >>>>> > Instead of FileBasedSink, build on FileIO.write; instead of >> >>>>> > FileBasedReader, build on FileIO.read. >> >>>>> >> >>>>> +1 >> >>>>> >> >>>>> I think the sink could be written atop FileIO.write, possibly using >> >>>>> dynamic destinations. At the very least the FileSink interface, which >> >>>>> handles the details of writing a single shard, would be an ideal way >> >>>>> to parameterize an SMB sink. It seems that none of our existing IOs >> >>>>> (publically?) expose FileSink implementations. >> >>>>> >> >>>>> FileIO.read is not flexible enough to do the merging. Eugene, is there >> >>>>> a composable analogue to FileSink, for sources, i.e. something that >> >>>>> can turn a file handle (possibly with offsets) into a set of records >> >>>>> other than FileBasedReader? >> >>>>> >> >>>>> > On Mon, Jul 15, 2019 at 9:01 AM Gleb Kanterov <[email protected]> >> >>>>> > wrote: >> >>>>> >> >> >>>>> >> I share the same concern with Robert regarding re-implementing >> >>>>> >> parts of IO. At the same time, in the past, I worked on internal >> >>>>> >> libraries that try to re-use code from existing IO, and it's hardly >> >>>>> >> possible because it feels like it wasn't designed for re-use. There >> >>>>> >> are a lot of classes that are nested (non-static) or non-public. I >> >>>>> >> can understand why they were made non-public, it's a hard >> >>>>> >> abstraction to design well and keep compatibility. As Neville >> >>>>> >> mentioned, decoupling readers and writers would not only benefit >> >>>>> >> for this proposal but for any other use-case that has to deal with >> >>>>> >> low-level API such as FileSystem API, that is hardly possible today >> >>>>> >> without copy-pasting, >> >>>>> >> >> >>>>> >> >> >>>>> >> >> >>>>> >> >> >>>>> >> >> >>>>> >> On Mon, Jul 15, 2019 at 5:05 PM Neville Li <[email protected]> >> >>>>> >> wrote: >> >>>>> >>> >> >>>>> >>> Re: avoiding mirroring IO functionality, what about: >> >>>>> >>> >> >>>>> >>> - Decouple the nested FileBasedSink.Writer and >> >>>>> >>> FileBasedSource.FileBasedReader, make them top level and remove >> >>>>> >>> references to parent classes. >> >>>>> >>> - Simplify the interfaces, while maintaining support for >> >>>>> >>> block/offset read & sequential write. >> >>>>> >>> - As a bonus, the refactored IO classes can be used standalone in >> >>>>> >>> case when the user wants to perform custom IO in a DoFn, i.e. a >> >>>>> >>> PTransform<PCollection<URI>, PCollection<KV<URI, GenericRecord>>>. >> >>>>> >>> Today this requires a lot of copy-pasted Avro boilerplate. >> >>>>> >>> - For compatibility, we can delegate to the new classes from the >> >>>>> >>> old ones and remove them in the next breaking release. >> >>>>> >>> >> >>>>> >>> Re: WriteFiles logic, I'm not sure about generalizing it, but what >> >>>>> >>> about splitting the part handling writing temp files into a new >> >>>>> >>> PTransform<PCollection<KV<ResourceId, Iterable<UserT>>>, >> >>>>> >>> PCollection<WriteFilesResult<DestinationT>>>? That splits the >> >>>>> >>> bucket-shard logic from actual file IO. >> >>>>> >>> >> >>>>> >>> On Mon, Jul 15, 2019 at 10:27 AM Robert Bradshaw >> >>>>> >>> <[email protected]> wrote: >> >>>>> >>>> >> >>>>> >>>> I agree that generalizing the existing FileIO may not be the right >> >>>>> >>>> path forward, and I'd only make their innards public with great >> >>>>> >>>> care. >> >>>>> >>>> (Would this be used like like >> >>>>> >>>> SmbSink(MyFileIO.sink(parameters).getWriter[Factory]())?) SMB is >> >>>>> >>>> a bit >> >>>>> >>>> unique that the source and sink are much more coupled than other >> >>>>> >>>> sources and sinks (which happen to be completely independent, if >> >>>>> >>>> complementary implementations, whereas SMB attempts to be a kind >> >>>>> >>>> of >> >>>>> >>>> pipe where one half is instanciated in each pipeline). >> >>>>> >>>> >> >>>>> >>>> In short, an SMB source/sink that is parameterized by an >> >>>>> >>>> arbitrary, >> >>>>> >>>> existing IO would be ideal (but possibly not feasible (per >> >>>>> >>>> existing >> >>>>> >>>> prioritizations)), or an SMB source/sink that works as a pair. >> >>>>> >>>> What >> >>>>> >>>> I'd like to avoid is a set of parallel SMB IO classes that >> >>>>> >>>> (partially, >> >>>>> >>>> and incompletely) mirror the existing IO ones (from an API >> >>>>> >>>> perspective--how much implementation it makes sense to share is an >> >>>>> >>>> orthogonal issue that I'm sure can be worked out.) >> >>>>> >>>> >> >>>>> >>>> On Mon, Jul 15, 2019 at 4:18 PM Neville Li >> >>>>> >>>> <[email protected]> wrote: >> >>>>> >>>> > >> >>>>> >>>> > Hi Robert, >> >>>>> >>>> > >> >>>>> >>>> > I agree, it'd be nice to reuse FileIO logic of different file >> >>>>> >>>> > types. But given the current code structure of FileIO & scope >> >>>>> >>>> > of the change, I feel it's better left for future refactor PRs. >> >>>>> >>>> > >> >>>>> >>>> > Some thoughts: >> >>>>> >>>> > - SMB file operation is simple single file sequential >> >>>>> >>>> > reads/writes, which already exists as Writer & FileBasedReader >> >>>>> >>>> > but are private inner classes, and have references to the >> >>>>> >>>> > parent Sink/Source instance. >> >>>>> >>>> > - The readers also have extra offset/split logic but that can >> >>>>> >>>> > be worked around. >> >>>>> >>>> > - It'll be nice to not duplicate temp->destination file logic >> >>>>> >>>> > but again WriteFiles is assuming a single integer shard key, so >> >>>>> >>>> > it'll take some refactoring to reuse it. >> >>>>> >>>> > >> >>>>> >>>> > All of these can be done in backwards compatible way. OTOH >> >>>>> >>>> > generalizing the existing components too much (esp. WriteFiles, >> >>>>> >>>> > which is already complex) might lead to two logic paths, one >> >>>>> >>>> > specialized for the SMB case. It might be easier to decouple >> >>>>> >>>> > some of them for better reuse. But again I feel it's a separate >> >>>>> >>>> > discussion. >> >>>>> >>>> > >> >>>>> >>>> > On Mon, Jul 15, 2019 at 9:45 AM Claire McGinty >> >>>>> >>>> > <[email protected]> wrote: >> >>>>> >>>> >> >> >>>>> >>>> >> Thanks Robert! >> >>>>> >>>> >> >> >>>>> >>>> >> We'd definitely like to be able to re-use existing I/O >> >>>>> >>>> >> components--for example the Writer<DestinationT, >> >>>>> >>>> >> OutputT>/FileBasedReader<T> (since they operate on a >> >>>>> >>>> >> WritableByteChannel/ReadableByteChannel, which is the level of >> >>>>> >>>> >> granularity we need) but the Writers, at least, seem to be >> >>>>> >>>> >> mostly private-access. Do you foresee them being made public >> >>>>> >>>> >> at any point? >> >>>>> >>>> >> >> >>>>> >>>> >> - Claire >> >>>>> >>>> >> >> >>>>> >>>> >> On Mon, Jul 15, 2019 at 9:31 AM Robert Bradshaw >> >>>>> >>>> >> <[email protected]> wrote: >> >>>>> >>>> >>> >> >>>>> >>>> >>> I left some comments on the doc. >> >>>>> >>>> >>> >> >>>>> >>>> >>> I think the general idea is sound, but one thing that worries >> >>>>> >>>> >>> me is >> >>>>> >>>> >>> the introduction of a parallel set of IOs that mirrors the >> >>>>> >>>> >>> (existing) >> >>>>> >>>> >>> FileIOs. I would suggest either (1) incorporate this >> >>>>> >>>> >>> functionality >> >>>>> >>>> >>> into the generic FileIO infrastructure, or let it be >> >>>>> >>>> >>> parameterized by >> >>>>> >>>> >>> arbitrary IO (which I'm not sure is possible, especially for >> >>>>> >>>> >>> the Read >> >>>>> >>>> >>> side (and better would be the capability of supporting >> >>>>> >>>> >>> arbitrary >> >>>>> >>>> >>> sources, aka an optional "as-sharded-source" operation that >> >>>>> >>>> >>> returns a >> >>>>> >>>> >>> PTransform<..., KV<shard-id, Iterable<KV<K, V>>> where the >> >>>>> >>>> >>> iterable is >> >>>>> >>>> >>> promised to be in key order)) or support a single SMB aka >> >>>>> >>>> >>> "PreGrouping" source/sink pair that's aways used together >> >>>>> >>>> >>> (and whose >> >>>>> >>>> >>> underlying format is not necessarily public). >> >>>>> >>>> >>> >> >>>>> >>>> >>> On Sat, Jul 13, 2019 at 3:19 PM Neville Li >> >>>>> >>>> >>> <[email protected]> wrote: >> >>>>> >>>> >>> > >> >>>>> >>>> >>> > 4 people have commented but mostly clarifying details and >> >>>>> >>>> >>> > not much on the overall design. >> >>>>> >>>> >>> > >> >>>>> >>>> >>> > It'd be great to have thumbs up/down on the design, >> >>>>> >>>> >>> > specifically metadata, bucket & shard strategy, etc., since >> >>>>> >>>> >>> > that affects backwards compatibility of output files. >> >>>>> >>>> >>> > Some breaking changes, e.g. dynamic # of shards, are out of >> >>>>> >>>> >>> > scope for V1 unless someone feels strongly about it. The >> >>>>> >>>> >>> > current scope should cover all our use cases and leave room >> >>>>> >>>> >>> > for optimization. >> >>>>> >>>> >>> > >> >>>>> >>>> >>> > Once green lighted we can start adopting internally, >> >>>>> >>>> >>> > ironing out rough edges while iterating on the PRs in >> >>>>> >>>> >>> > parallel. >> >>>>> >>>> >>> > >> >>>>> >>>> >>> > Most of the implementation is self-contained in the >> >>>>> >>>> >>> > extensions:smb module, except making a few core >> >>>>> >>>> >>> > classes/methods public for reuse. So despite the amount of >> >>>>> >>>> >>> > work it's still fairly low risk to the code base. There're >> >>>>> >>>> >>> > some proposed optimization & refactoring involving core >> >>>>> >>>> >>> > (see appendix) but IMO they're better left for followup PRs. >> >>>>> >>>> >>> > >> >>>>> >>>> >>> > On Fri, Jul 12, 2019 at 11:34 PM Kenneth Knowles >> >>>>> >>>> >>> > <[email protected]> wrote: >> >>>>> >>>> >>> >> >> >>>>> >>>> >>> >> I've seen some discussion on the doc. I cannot tell >> >>>>> >>>> >>> >> whether the questions are resolved or what the status of >> >>>>> >>>> >>> >> review is. Would you mind looping this thread with a quick >> >>>>> >>>> >>> >> summary? This is such a major piece of work I don't want >> >>>>> >>>> >>> >> it to sit with everyone thinking they are waiting on >> >>>>> >>>> >>> >> someone else, or any such thing. (not saying this is >> >>>>> >>>> >>> >> happening, just pinging to be sure) >> >>>>> >>>> >>> >> >> >>>>> >>>> >>> >> Kenn >> >>>>> >>>> >>> >> >> >>>>> >>>> >>> >> On Mon, Jul 1, 2019 at 1:09 PM Neville Li >> >>>>> >>>> >>> >> <[email protected]> wrote: >> >>>>> >>>> >>> >>> >> >>>>> >>>> >>> >>> Updated the doc a bit with more future work (appendix). >> >>>>> >>>> >>> >>> IMO most of them are non-breaking and better done in >> >>>>> >>>> >>> >>> separate PRs later since some involve pretty big >> >>>>> >>>> >>> >>> refactoring and are outside the scope of MVP. >> >>>>> >>>> >>> >>> >> >>>>> >>>> >>> >>> For now we'd really like to get feedback on some >> >>>>> >>>> >>> >>> fundamental design decisions and find a way to move >> >>>>> >>>> >>> >>> forward. >> >>>>> >>>> >>> >>> >> >>>>> >>>> >>> >>> On Thu, Jun 27, 2019 at 4:39 PM Neville Li >> >>>>> >>>> >>> >>> <[email protected]> wrote: >> >>>>> >>>> >>> >>>> >> >>>>> >>>> >>> >>>> Thanks. I responded to comments in the doc. More inline. >> >>>>> >>>> >>> >>>> >> >>>>> >>>> >>> >>>> On Thu, Jun 27, 2019 at 2:44 PM Chamikara Jayalath >> >>>>> >>>> >>> >>>> <[email protected]> wrote: >> >>>>> >>>> >>> >>>>> >> >>>>> >>>> >>> >>>>> Thanks added few comments. >> >>>>> >>>> >>> >>>>> >> >>>>> >>>> >>> >>>>> If I understood correctly, you basically assign >> >>>>> >>>> >>> >>>>> elements with keys to different buckets which are >> >>>>> >>>> >>> >>>>> written to unique files and merge files for the same >> >>>>> >>>> >>> >>>>> key while reading ? >> >>>>> >>>> >>> >>>>> >> >>>>> >>>> >>> >>>>> Some of my concerns are. >> >>>>> >>>> >>> >>>>> >> >>>>> >>>> >>> >>>>> (1) Seems like you rely on an in-memory sorting of >> >>>>> >>>> >>> >>>>> buckets. Will this end up limiting the size of a >> >>>>> >>>> >>> >>>>> PCollection you can process ? >> >>>>> >>>> >>> >>>> >> >>>>> >>>> >>> >>>> The sorter transform we're using supports spilling and >> >>>>> >>>> >>> >>>> external sort. We can break up large key groups further >> >>>>> >>>> >>> >>>> by sharding, similar to fan out in some GBK transforms. >> >>>>> >>>> >>> >>>> >> >>>>> >>>> >>> >>>>> (2) Seems like you rely on Reshuffle.viaRandomKey() >> >>>>> >>>> >>> >>>>> which is actually implemented using a shuffle (which >> >>>>> >>>> >>> >>>>> you try to replace with this proposal). >> >>>>> >>>> >>> >>>> >> >>>>> >>>> >>> >>>> That's for distributing task metadata, so that each DoFn >> >>>>> >>>> >>> >>>> thread picks up a random bucket and sort merge >> >>>>> >>>> >>> >>>> key-values. It's not shuffling actual data. >> >>>>> >>>> >>> >>>> >> >>>>> >>>> >>> >>>>> >> >>>>> >>>> >>> >>>>> (3) I think (at least some of the) shuffle >> >>>>> >>>> >>> >>>>> implementations are implemented in ways similar to this >> >>>>> >>>> >>> >>>>> (writing to files and merging). So I'm wondering if the >> >>>>> >>>> >>> >>>>> performance benefits you see are for a very specific >> >>>>> >>>> >>> >>>>> case and may limit the functionality in other ways. >> >>>>> >>>> >>> >>>> >> >>>>> >>>> >>> >>>> This is for the common pattern of few core data producer >> >>>>> >>>> >>> >>>> pipelines and many downstream consumer pipelines. It's >> >>>>> >>>> >>> >>>> not intended to replace shuffle/join within a single >> >>>>> >>>> >>> >>>> pipeline. On the producer side, by pre-grouping/sorting >> >>>>> >>>> >>> >>>> data and writing to bucket/shard output files, the >> >>>>> >>>> >>> >>>> consumer can sort/merge matching ones without a CoGBK. >> >>>>> >>>> >>> >>>> Essentially we're paying the shuffle cost upfront to >> >>>>> >>>> >>> >>>> avoid them repeatedly in each consumer pipeline that >> >>>>> >>>> >>> >>>> wants to join data. >> >>>>> >>>> >>> >>>> >> >>>>> >>>> >>> >>>>> >> >>>>> >>>> >>> >>>>> Thanks, >> >>>>> >>>> >>> >>>>> Cham >> >>>>> >>>> >>> >>>>> >> >>>>> >>>> >>> >>>>> >> >>>>> >>>> >>> >>>>> On Thu, Jun 27, 2019 at 8:12 AM Neville Li >> >>>>> >>>> >>> >>>>> <[email protected]> wrote: >> >>>>> >>>> >>> >>>>>> >> >>>>> >>>> >>> >>>>>> Ping again. Any chance someone takes a look to get >> >>>>> >>>> >>> >>>>>> this thing going? It's just a design doc and basic >> >>>>> >>>> >>> >>>>>> metadata/IO impl. We're not talking about actual >> >>>>> >>>> >>> >>>>>> source/sink code yet (already done but saved for >> >>>>> >>>> >>> >>>>>> future PRs). >> >>>>> >>>> >>> >>>>>> >> >>>>> >>>> >>> >>>>>> On Fri, Jun 21, 2019 at 1:38 PM Ahmet Altay >> >>>>> >>>> >>> >>>>>> <[email protected]> wrote: >> >>>>> >>>> >>> >>>>>>> >> >>>>> >>>> >>> >>>>>>> Thank you Claire, this looks promising. Explicitly >> >>>>> >>>> >>> >>>>>>> adding a few folks that might have feedback: +Ismaël >> >>>>> >>>> >>> >>>>>>> Mejía +Robert Bradshaw +Lukasz Cwik +Chamikara >> >>>>> >>>> >>> >>>>>>> Jayalath >> >>>>> >>>> >>> >>>>>>> >> >>>>> >>>> >>> >>>>>>> On Mon, Jun 17, 2019 at 2:12 PM Claire McGinty >> >>>>> >>>> >>> >>>>>>> <[email protected]> wrote: >> >>>>> >>>> >>> >>>>>>>> >> >>>>> >>>> >>> >>>>>>>> Hey dev@! >> >>>>> >>>> >>> >>>>>>>> >> >>>>> >>>> >>> >>>>>>>> Myself and a few other Spotify data engineers have >> >>>>> >>>> >>> >>>>>>>> put together a design doc for SMB Join support in >> >>>>> >>>> >>> >>>>>>>> Beam, and have a working Java implementation we've >> >>>>> >>>> >>> >>>>>>>> started to put up for PR ([0], [1], [2]). There's >> >>>>> >>>> >>> >>>>>>>> more detailed information in the document, but the >> >>>>> >>>> >>> >>>>>>>> tl;dr is that SMB is a strategy to optimize joins >> >>>>> >>>> >>> >>>>>>>> for file-based sources by modifying the initial >> >>>>> >>>> >>> >>>>>>>> write operation to write records in sorted buckets >> >>>>> >>>> >>> >>>>>>>> based on the desired join key. This means that >> >>>>> >>>> >>> >>>>>>>> subsequent joins of datasets written in this way are >> >>>>> >>>> >>> >>>>>>>> only sequential file reads, no shuffling involved. >> >>>>> >>>> >>> >>>>>>>> We've seen some pretty substantial performance >> >>>>> >>>> >>> >>>>>>>> speedups with our implementation and would love to >> >>>>> >>>> >>> >>>>>>>> get it checked in to Beam's Java SDK. >> >>>>> >>>> >>> >>>>>>>> >> >>>>> >>>> >>> >>>>>>>> We'd appreciate any suggestions or feedback on our >> >>>>> >>>> >>> >>>>>>>> proposal--the design doc should be public to comment >> >>>>> >>>> >>> >>>>>>>> on. >> >>>>> >>>> >>> >>>>>>>> >> >>>>> >>>> >>> >>>>>>>> Thanks! >> >>>>> >>>> >>> >>>>>>>> Claire / Neville >> >>>>> >> >> >>>>> >> >> >>>>> >> >> >>>>> >> -- >> >>>>> >> Cheers, >> >>>>> >> Gleb >> > >> > >> > >> > -- >> > Cheers, >> > Gleb > > > > -- > Cheers, > Gleb
