TextIO.read() and AvroIO.read() indeed perform better than match() + readMatches() + readFiles(), due to DWR - so for these two in particular I would not recommend such a refactoring. However, new file-based IOs that do not support DWR should only provide readFiles(). Those that do, should provide read() and readFiles(). When SDF supports DWR, then readFiles() will be enough in all cases. In general there's no need for readAll() for new file-based IOs - it is always equivalent to matchAll() + readMatches() + readFiles() including performance-wise. It was included in TextIO/AvroIO before readFiles() was a thing.
On Wed, Jan 30, 2019 at 2:41 PM Chamikara Jayalath <[email protected]> wrote: > On Wed, Jan 30, 2019 at 2:37 PM Chamikara Jayalath <[email protected]> > wrote: > >> >> >> On Wed, Jan 30, 2019 at 2:33 PM Ismaël Mejía <[email protected]> wrote: >> >>> Ups slight typo, in the first line of the previous email I meant read >>> instead of readAll >>> >>> On Wed, Jan 30, 2019 at 11:32 PM Ismaël Mejía <[email protected]> wrote: >>> > >>> > Reuven is right for the example, readAll at this moment may be faster >>> > and also supports Dynamic Work Rebalancing (DWR), but the performance >>> > of the other approach may (and must) be improved to be equal, once the >>> > internal implementation of TextIO.read moves to a SDF version instead >>> > of the FileBasedSource one, and once that runners support DWR through >>> > SDF. Of course all of this is future work. Probably Eugene can >>> > eventually chime in to give more details in practical performance in >>> > his tests in Dataflow. >>> > >>> > Really interesting topic, but I want to bring back the discussion to >>> > the subject of the thread. I think there is some confusion after >>> > Jeff's example which should have been: >>> > >>> > return input >>> > .apply(TextIO.readAll()); >>> > >>> > to: >>> > >>> > return input >>> > .apply(FileIO.match().filepattern(fileSpec)) >>> > .apply(FileIO.readMatches()) >>> > .apply(TextIO.readFiles()); >>> > >>> > This is the question we are addressing, do we need a readAll transform >>> > that replaces the 3 steps or no? >>> >> >> Ismaël, I'm not quite sure how these two are equal. readFiles() transform >> returns a PCollection of ReadableFile objects. Users are expected to read >> these files in a subsequent ParDo and produce a PCollection of proper type. >> FooIO.ReadAll() transforms on the other hand are tailored to each IO >> connector and return a PCollection of objects of type that are supported to >> be returned by that IO connector. >> > > I assume you meant FileIO.readFiles() here. Or did you mean > TextIO.readFiles() ? If so that seems very similar to TextIO.readAll(). > >> >> >> >>> > >>> > On Wed, Jan 30, 2019 at 9:03 PM Robert Bradshaw <[email protected]> >>> wrote: >>> > > >>> > > Yes, this is precisely the goal of SDF. >>> > > >>> > > >>> > > On Wed, Jan 30, 2019 at 8:41 PM Kenneth Knowles <[email protected]> >>> wrote: >>> > > > >>> > > > So is the latter is intended for splittable DoFn but not yet using >>> it? The promise of SDF is precisely this composability, isn't it? >>> > > > >>> > > > Kenn >>> > > > >>> > > > On Wed, Jan 30, 2019 at 10:16 AM Jeff Klukas <[email protected]> >>> wrote: >>> > > >> >>> > > >> Reuven - Is TextIO.read().from() a more complex case than the >>> topic Ismaël is bringing up in this thread? I'm surprised to hear that the >>> two examples have different performance characteristics. >>> > > >> >>> > > >> Reading through the implementation, I guess the fundamental >>> difference is whether a given configuration expands to TextIO.ReadAll or to >>> io.Read. AFAICT, that detail and the subsequent performance impact is not >>> documented. >>> > > >> >>> > > >> If the above is correct, perhaps it's an argument for IOs to >>> provide higher-level methods in cases where they can optimize performance >>> compared to what a user might naively put together. >>> > > >> >>> > > >> On Wed, Jan 30, 2019 at 12:35 PM Reuven Lax <[email protected]> >>> wrote: >>> > > >>> >>> > > >>> Jeff, what you did here is not simply a refactoring. These two >>> are quite different, and will likely have different performance >>> characteristics. >>> > > >>> >>> > > >>> The first evaluates the wildcard, and allows the runner to pick >>> appropriate bundling. Bundles might contain multiple files (if they are >>> small), and the runner can split the files as appropriate. In the case of >>> the Dataflow runner, these bundles can be further split dynamically. >>> > > >>> >>> > > >>> The second chops of the files inside the the PTransform, and >>> processes each chunk in a ParDo. TextIO.readFiles currently chops up each >>> file into 64mb chunks (hardcoded), and then processes each chunk in a ParDo. >>> > > >>> >>> > > >>> Reuven >>> > > >>> >>> > > >>> >>> > > >>> On Wed, Jan 30, 2019 at 9:18 AM Jeff Klukas <[email protected]> >>> wrote: >>> > > >>>> >>> > > >>>> I would prefer we move towards option [2]. I just tried the >>> following refactor in my own code from: >>> > > >>>> >>> > > >>>> return input >>> > > >>>> .apply(TextIO.read().from(fileSpec)); >>> > > >>>> >>> > > >>>> to: >>> > > >>>> >>> > > >>>> return input >>> > > >>>> .apply(FileIO.match().filepattern(fileSpec)) >>> > > >>>> .apply(FileIO.readMatches()) >>> > > >>>> .apply(TextIO.readFiles()); >>> > > >>>> >>> > > >>>> Yes, the latter is more verbose but not ridiculously so, and >>> it's also more instructive about what's happening. >>> > > >>>> >>> > > >>>> When I first started working with Beam, it took me a while to >>> realize that TextIO.read().from() would accept a wildcard. The more verbose >>> version involves a method called "filepattern" which makes this much more >>> obvious. It also leads me to understand that I could use the same >>> FileIO.match() machinery to do other things with filesystems other than >>> read file contents. >>> > > >>>> >>> > > >>>> On Wed, Jan 30, 2019 at 11:26 AM Ismaël Mejía < >>> [email protected]> wrote: >>> > > >>>>> >>> > > >>>>> Hello, >>> > > >>>>> >>> > > >>>>> A ‘recent’ pattern of use in Beam is to have in file based IOs >>> a >>> > > >>>>> `readAll()` implementation that basically matches a >>> `PCollection` of >>> > > >>>>> file patterns and reads them, e.g. `TextIO`, `AvroIO`. >>> `ReadAll` is >>> > > >>>>> implemented by a expand function that matches files with >>> FileIO and >>> > > >>>>> then reads them using a format specific `ReadFiles` transform >>> e.g. >>> > > >>>>> TextIO.ReadFiles, AvroIO.ReadFiles. So in the end `ReadAll` in >>> the >>> > > >>>>> Java implementation is just an user friendly API to hide >>> FileIO.match >>> > > >>>>> + ReadFiles. >>> > > >>>>> >>> > > >>>>> Most recent IOs do NOT implement ReadAll to encourage the more >>> > > >>>>> composable approach of File + ReadFiles, e.g. XmlIO and >>> ParquetIO. >>> > > >>>>> >>> > > >>>>> Implementing ReadAll as a wrapper is relatively easy and is >>> definitely >>> > > >>>>> user friendly, but it has an issue, it may be error-prone and >>> it adds >>> > > >>>>> more code to maintain (mostly ‘repeated’ code). However >>> `readAll` is a >>> > > >>>>> more abstract pattern that applies not only to File based IOs >>> so it >>> > > >>>>> makes sense for example in other transforms that map a >>> `Pcollection` >>> > > >>>>> of read requests and is the basis for SDF composable style >>> APIs like >>> > > >>>>> the recent `HBaseIO.readAll()`. >>> > > >>>>> >>> > > >>>>> So the question is should we: >>> > > >>>>> >>> > > >>>>> [1] Implement `readAll` in all file based IOs to be user >>> friendly and >>> > > >>>>> assume the (minor) maintenance cost >>> > > >>>>> >>> > > >>>>> or >>> > > >>>>> >>> > > >>>>> [2] Deprecate `readAll` from file based IOs and encourage >>> users to use >>> > > >>>>> FileIO + `readFiles` (less maintenance and encourage >>> composition). >>> > > >>>>> >>> > > >>>>> I just checked quickly in the python code base but I did not >>> find if >>> > > >>>>> the File match + ReadFiles pattern applies, but it would be >>> nice to >>> > > >>>>> see what the python guys think on this too. >>> > > >>>>> >>> > > >>>>> This discussion comes from a recent slack conversation with >>> Łukasz >>> > > >>>>> Gajowy, and we wanted to settle into one approach to make the >>> IO >>> > > >>>>> signatures consistent, so any opinions/preferences? >>> >>
