higher dplyr interface, extra sdf_* features, and RDS-based serialization routines

We’re thrilled to announce sparklyr 1.5 is now
out there on CRAN!

To put in sparklyr 1.5 from CRAN, run

On this weblog put up, we’ll spotlight the next points of sparklyr 1.5:

Higher dplyr interface

A big fraction of pull requests that went into the sparklyr 1.5 launch had been targeted on making
Spark dataframes work with numerous dplyr verbs in the identical manner that R dataframes do.
The complete listing of dplyr-related bugs and have requests that had been resolved in
sparklyr 1.5 might be present in right here.

On this part, we’ll showcase three new dplyr functionalities that had been shipped with sparklyr 1.5.

Stratified sampling

Stratified sampling on an R dataframe might be achieved with a mix of dplyr::group_by() adopted by
dplyr::sample_n() or dplyr::sample_frac(), the place the grouping variables specified within the dplyr::group_by()
step are those that outline every stratum. As an example, the next question will group mtcars by quantity
of cylinders and return a weighted random pattern of dimension two from every group, with out alternative, and weighted by
the mpg column:

## # A tibble: 6 x 11
## # Teams:   cyl [3]
##     mpg   cyl  disp    hp  drat    wt  qsec    vs    am  gear  carb
##             
## 1  33.9     4  71.1    65  4.22  1.84  19.9     1     1     4     1
## 2  22.8     4 108      93  3.85  2.32  18.6     1     1     4     1
## 3  21.4     6 258     110  3.08  3.22  19.4     1     0     3     1
## 4  21       6 160     110  3.9   2.62  16.5     0     1     4     4
## 5  15.5     8 318     150  2.76  3.52  16.9     0     0     3     2
## 6  19.2     8 400     175  3.08  3.84  17.0     0     0     3     2

Ranging from sparklyr 1.5, the identical can be achieved for Spark dataframes with Spark 3.0 or above, e.g.,:

library(sparklyr)

sc <- spark_connect(grasp = "native", model = "3.0.0")
mtcars_sdf <- copy_to(sc, mtcars, exchange = TRUE, repartition = 3)

mtcars_sdf %>%
  dplyr::group_by(cyl) %>%
  dplyr::sample_n(dimension = 2, weight = mpg, exchange = FALSE) %>%
  print()

# Supply: spark> [?? x 11]
# Teams: cyl
    mpg   cyl  disp    hp  drat    wt  qsec    vs    am  gear  carb
            
1  21       6 160     110  3.9   2.62  16.5     0     1     4     4
2  21.4     6 258     110  3.08  3.22  19.4     1     0     3     1
3  27.3     4  79      66  4.08  1.94  18.9     1     1     4     1
4  32.4     4  78.7    66  4.08  2.2   19.5     1     1     4     1
5  16.4     8 276.    180  3.07  4.07  17.4     0     0     3     3
6  18.7     8 360     175  3.15  3.44  17.0     0     0     3     2

or

## # Supply: spark> [?? x 11]
## # Teams: cyl
##     mpg   cyl  disp    hp  drat    wt  qsec    vs    am  gear  carb
##             
## 1  21       6 160     110  3.9   2.62  16.5     0     1     4     4
## 2  21.4     6 258     110  3.08  3.22  19.4     1     0     3     1
## 3  22.8     4 141.     95  3.92  3.15  22.9     1     0     4     2
## 4  33.9     4  71.1    65  4.22  1.84  19.9     1     1     4     1
## 5  30.4     4  95.1   113  3.77  1.51  16.9     1     1     5     2
## 6  15.5     8 318     150  2.76  3.52  16.9     0     0     3     2
## 7  18.7     8 360     175  3.15  3.44  17.0     0     0     3     2
## 8  16.4     8 276.    180  3.07  4.07  17.4     0     0     3     3

Row sums

The rowSums() performance supplied by dplyr is useful when one must sum up
a lot of columns inside an R dataframe which might be impractical to be enumerated
individually.
For instance, right here now we have a six-column dataframe of random actual numbers, the place the
partial_sum column within the outcome accommodates the sum of columns b by means of d inside
every row:

## # A tibble: 5 x 7
##         a     b     c      d     e      f partial_sum
##                   
## 1 0.781   0.801 0.157 0.0293 0.169 0.0978        1.16
## 2 0.696   0.412 0.221 0.941  0.697 0.675         2.27
## 3 0.802   0.410 0.516 0.923  0.190 0.904         2.04
## 4 0.200   0.590 0.755 0.494  0.273 0.807         2.11
## 5 0.00149 0.711 0.286 0.297  0.107 0.425         1.40

Starting with sparklyr 1.5, the identical operation might be carried out with Spark dataframes:

## # Supply: spark> [?? x 7]
##         a     b     c      d     e      f partial_sum
##                   
## 1 0.781   0.801 0.157 0.0293 0.169 0.0978        1.16
## 2 0.696   0.412 0.221 0.941  0.697 0.675         2.27
## 3 0.802   0.410 0.516 0.923  0.190 0.904         2.04
## 4 0.200   0.590 0.755 0.494  0.273 0.807         2.11
## 5 0.00149 0.711 0.286 0.297  0.107 0.425         1.40

As a bonus from implementing the rowSums function for Spark dataframes,
sparklyr 1.5 now additionally provides restricted assist for the column-subsetting
operator on Spark dataframes.
For instance, all code snippets beneath will return some subset of columns from
the dataframe named sdf:

# choose columns `b` by means of `e`
sdf[2:5]

# choose columns `b` and `c`
sdf[c("b", "c")]

# drop the primary and third columns and return the remainder
sdf[c(-1, -3)]

Weighted-mean summarizer

Much like the 2 dplyr features talked about above, the weighted.imply() summarizer is one other
helpful perform that has turn out to be a part of the dplyr interface for Spark dataframes in sparklyr 1.5.
One can see it in motion by, for instance, evaluating the output from the next

with output from the equal operation on mtcars in R:

each of them ought to consider to the next:

##     cyl mpg_wm
##     
## 1     4   25.9
## 2     6   19.6
## 3     8   14.8

New additions to the sdf_* household of features

sparklyr supplies a lot of comfort features for working with Spark dataframes,
and all of them have names beginning with the sdf_ prefix.

On this part we’ll briefly point out 4 new additions
and present some instance situations by which these features are helpful.

sdf_expand_grid()

Because the title suggests, sdf_expand_grid() is just the Spark equal of broaden.grid().
Slightly than operating broaden.grid() in R and importing the ensuing R dataframe to Spark, one
can now run sdf_expand_grid(), which accepts each R vectors and Spark dataframes and helps
hints for broadcast hash joins. The instance beneath exhibits sdf_expand_grid() making a
100-by-100-by-10-by-10 grid in Spark over 1000 Spark partitions, with broadcast hash be part of hints
on variables with small cardinalities:

library(sparklyr)

sc <- spark_connect(grasp = "native")

grid_sdf <- sdf_expand_grid(
  sc,
  var1 = seq(100),
  var2 = seq(100),
  var3 = seq(10),
  var4 = seq(10),
  broadcast_vars = c(var3, var4),
  repartition = 1000
)

grid_sdf %>% sdf_nrow() %>% print()

## [1] 1e+06

sdf_partition_sizes()

As sparklyr person @sbottelli urged right here,
one factor that might be nice to have in sparklyr is an environment friendly approach to question partition sizes of a Spark dataframe.
In sparklyr 1.5, sdf_partition_sizes() does precisely that:

library(sparklyr)

sc <- spark_connect(grasp = "native")

sdf_len(sc, 1000, repartition = 5) %>%
  sdf_partition_sizes() %>%
  print(row.names = FALSE)

##  partition_index partition_size
##                0            200
##                1            200
##                2            200
##                3            200
##                4            200

sdf_unnest_longer() and sdf_unnest_wider()

sdf_unnest_longer() and sdf_unnest_wider() are the equivalents of
tidyr::unnest_longer() and tidyr::unnest_wider() for Spark dataframes.
sdf_unnest_longer() expands all components in a struct column into a number of rows, and
sdf_unnest_wider() expands them into a number of columns. As illustrated with an instance
dataframe beneath,

library(sparklyr)

sc <- spark_connect(grasp = "native")
sdf <- copy_to(
  sc,
  tibble::tibble(
    id = seq(3),
    attribute = listing(
      listing(title = "Alice", grade = "A"),
      listing(title = "Bob", grade = "B"),
      listing(title = "Carol", grade = "C")
    )
  )
)

sdf %>%
  sdf_unnest_longer(col = document, indices_to = "key", values_to = "worth") %>%
  print()

evaluates to

## # Supply: spark> [?? x 3]
##      id worth key
##     
## 1     1 A     grade
## 2     1 Alice title
## 3     2 B     grade
## 4     2 Bob   title
## 5     3 C     grade
## 6     3 Carol title

whereas

sdf %>%
  sdf_unnest_wider(col = document) %>%
  print()

evaluates to

## # Supply: spark> [?? x 3]
##      id grade title
##     
## 1     1 A     Alice
## 2     2 B     Bob
## 3     3 C     Carol

RDS-based serialization routines

Some readers should be questioning why a model new serialization format would have to be carried out in sparklyr in any respect.
Lengthy story brief, the reason being that RDS serialization is a strictly higher alternative for its CSV predecessor.
It possesses all fascinating attributes the CSV format has,
whereas avoiding quite a lot of disadvantages which might be widespread amongst text-based knowledge codecs.

On this part, we’ll briefly define why sparklyr ought to assist no less than one serialization format aside from arrow,
deep-dive into points with CSV-based serialization,
after which present how the brand new RDS-based serialization is free from these points.

Why arrow shouldn’t be for everybody?

To switch knowledge between Spark and R accurately and effectively, sparklyr should depend on some knowledge serialization
format that’s well-supported by each Spark and R.
Sadly, not many serialization codecs fulfill this requirement,
and among the many ones that do are text-based codecs equivalent to CSV and JSON,
and binary codecs equivalent to Apache Arrow, Protobuf, and as of current, a small subset of RDS model 2.
Additional complicating the matter is the extra consideration that
sparklyr ought to assist no less than one serialization format whose implementation might be absolutely self-contained inside the sparklyr code base,
i.e., such serialization shouldn’t depend upon any exterior R package deal or system library,
in order that it could actually accommodate customers who wish to use sparklyr however who don’t essentially have the required C++ compiler instrument chain and
different system dependencies for establishing R packages equivalent to arrow or
protolite.
Previous to sparklyr 1.5, CSV-based serialization was the default different to fallback to when customers should not have the arrow package deal put in or
when the kind of knowledge being transported from R to Spark is unsupported by the model of arrow out there.

Why is the CSV format not perfect?

There are no less than three causes to imagine CSV format shouldn’t be the only option with regards to exporting knowledge from R to Spark.

One purpose is effectivity. For instance, a double-precision floating level quantity equivalent to .Machine$double.eps must
be expressed as "2.22044604925031e-16" in CSV format as a way to not incur any lack of precision, thus taking over 20 bytes
somewhat than 8 bytes.

However extra necessary than effectivity are correctness considerations. In a R dataframe, one can retailer each NA_real_ and
NaN in a column of floating level numbers. NA_real_ ought to ideally translate to null inside a Spark dataframe, whereas
NaN ought to proceed to be NaN when transported from R to Spark. Sadly, NA_real_ in R turns into indistinguishable
from NaN as soon as serialized in CSV format, as evident from a fast demo proven beneath:

##     x is_nan
## 1  NA  FALSE
## 2 NaN   TRUE

csv_file <- "/tmp/knowledge.csv"
write.csv(original_df, file = csv_file, row.names = FALSE)
deserialized_df <- learn.csv(csv_file)
deserialized_df %>% dplyr::mutate(is_nan = is.nan(x)) %>% print()

##    x is_nan
## 1 NA  FALSE
## 2 NA  FALSE

One other correctness concern very a lot just like the one above was the truth that
"NA" and NA inside a string column of an R dataframe turn out to be indistinguishable
as soon as serialized in CSV format, as accurately identified in
this Github concern
by @caewok and others.

RDS to the rescue!

RDS format is among the most generally used binary codecs for serializing R objects.
It’s described in some element in chapter 1, part 8 of
this doc.
Amongst benefits of the RDS format are effectivity and accuracy: it has a fairly
environment friendly implementation in base R, and helps all R knowledge varieties.

Additionally price noticing is the truth that when an R dataframe containing solely knowledge varieties
with wise equivalents in Apache Spark (e.g., RAWSXP, LGLSXP, CHARSXP, REALSXP, and many others)
is saved utilizing RDS model 2,
(e.g., serialize(mtcars, connection = NULL, model = 2L, xdr = TRUE)),
solely a tiny subset of the RDS format will likely be concerned within the serialization course of,
and implementing deserialization routines in Scala able to decoding such a restricted
subset of RDS constructs is in actual fact a fairly easy and easy job
(as proven in
right here
).

Final however not least, as a result of RDS is a binary format, it permits NA_character_, "NA",
NA_real_, and NaN to all be encoded in an unambiguous method, therefore permitting sparklyr
1.5 to keep away from all correctness points detailed above in non-arrow serialization use circumstances.

Different advantages of RDS serialization

Along with correctness ensures, RDS format additionally provides fairly just a few different benefits.

One benefit is after all efficiency: for instance, importing a non-trivially-sized dataset
equivalent to nycflights13::flights from R to Spark utilizing the RDS format in sparklyr 1.5 is
roughly 40%-50% sooner in comparison with CSV-based serialization in sparklyr 1.4. The
present RDS-based implementation remains to be nowhere as quick as arrow-based serialization
although (arrow is about 3-4x sooner), so for performance-sensitive duties involving
heavy serialization, arrow ought to nonetheless be the best choice.

One other benefit is that with RDS serialization, sparklyr can import R dataframes containing
uncooked columns immediately into binary columns in Spark. Thus, use circumstances such because the one beneath
will work in sparklyr 1.5

Whereas most sparklyr customers in all probability received’t discover this functionality of importing binary columns
to Spark instantly helpful of their typical sparklyr::copy_to() or sparklyr::acquire()
usages, it does play a vital function in lowering serialization overheads within the Spark-based
foreach parallel backend that
was first launched in sparklyr 1.2.
It is because Spark staff can immediately fetch the serialized R closures to be computed
from a binary Spark column as a substitute of extracting these serialized bytes from intermediate
representations equivalent to base64-encoded strings.
Equally, the R outcomes from executing employee closures will likely be immediately out there in RDS
format which might be effectively deserialized in R, somewhat than being delivered in different
much less environment friendly codecs.

Acknowledgement

In chronological order, we wish to thank the next contributors for making their pull
requests a part of sparklyr 1.5:

We might additionally like to precise our gratitude in the direction of quite a few bug stories and have requests for
sparklyr from a improbable open-source neighborhood.

Lastly, the writer of this weblog put up is indebted to
@javierluraschi,
@batpigandme,
and @skeydan for his or her worthwhile editorial inputs.

If you happen to want to be taught extra about sparklyr, try sparklyr.ai,
spark.rstudio.com, and among the earlier launch posts equivalent to
sparklyr 1.4 and
sparklyr 1.3.

Thanks for studying!