The ability to analyze time series data at scale is critical for the success of finance and IoT applications based on Spark. Flint is Two Sigma’s implementation of highly optimized time series operations in Spark. It performs truly parallel and rich analyses on time series data by taking advantage of the natural ordering in time series data to provide locality-based optimizations.

Flint is an open source library for Spark based around the TimeSeriesRDD, a time series aware data structure, and a collection of time series utility and analysis functions that use TimeSeriesRDDs. Unlike DataFrame and Dataset, Flint’s TimeSeriesRDDs can leverage the existing ordering properties of datasets at rest and the fact that almost all data manipulations and analysis over these datasets respect their temporal ordering properties. It differs from other time series efforts in Spark in its ability to efficiently compute across panel data or on large scale high frequency data.

Spark version: 2.0 and above

Scala version: 2.11.7 and above

Python version: 3.5 and above

To build this sbt project, one could simply do

sbt assembly

The python bindings for Flint, including quickstart instructions, are documented at python/README.md. API documentation is available at https://ts-flint.readthedocs.io/en/latest

The entry point into all functionalities for time series analysis in Flint is the TimeSeriesRDD class or object. In high level, a TimeSeriesRDD contains an OrderedRDD which could be used to represent a sequence of ordering key-value pairs. A TimeSeriesRDD uses Long to represent timestamps in nanoseconds since epoch as keys and InternalRows as values for OrderedRDD to represent a time series data set.

Applications can create a TimeSeriesRDD from an existing RDD, from an OrderedRDD, from a DataFrame, or from a single csv file.

As an example, the following creates a TimeSeriesRDD from a gzipped CSV file with header and specific datetime format.

import com.twosigma.flint.timeseries.CSV
val tsRdd = CSV.from(
  sqlContext,
  "file://foo/bar/data.csv",
  header = true,
  dateFormat = "yyyyMMdd HH:mm:ss.SSS",
  codec = "gzip",
  sorted = true
)

To create a TimeSeriesRDD from a DataFrame, you have to make sure the DataFrame contains a column named "time" of type LongType.

import com.twosigma.flint.timeseries.TimeSeriesRDD
import scala.concurrent.duration._
val df = ... // A DataFrame whose rows have been sorted by their timestamps under "time" column
val tsRdd = TimeSeriesRDD.fromDF(dataFrame = df)(isSorted = true, timeUnit = MILLISECONDS)

One could also create a TimeSeriesRDD from a RDD[Row] or an OrderedRDD[Long, Row] by providing a schema, e.g.

import com.twosigma.flint.timeseries._
import scala.concurrent.duration._
val rdd = ... // An RDD whose rows have sorted by their timestamps
val tsRdd = TimeSeriesRDD.fromRDD(
  rdd,
  schema = Schema("time" -> LongType, "price" -> DoubleType)
)(isSorted = true,
  timeUnit = MILLISECONDS
)

It is also possible to create a TimeSeriesRDD from a dataset stored as parquet format file(s). The TimeSeriesRDD.fromParquet() function provides the option to specify which columns and/or the time range you are interested, e.g.

import com.twosigma.flint.timeseries._
import scala.concurrent.duration._
val tsRdd = TimeSeriesRDD.fromParquet(
  sqlContext,
  path = "hdfs://foo/bar/"
)(isSorted = true,
  timeUnit = MILLISECONDS,
  columns = Seq("time", "id", "price"),  // By default, null for all columns
  begin = "20100101",                    // By default, null for no boundary at begin
  end = "20150101"                       // By default, null for no boundary at end
)

Similar to DataFrame, one could get the schema of a TimeSeriesRDD, and perform operations like first(), cache(), collect(), repartition(), persist(), etc. Other than those basic operations supported by DataFrame or RDD, one could manipulate rows and columns with the following functions.

• cast A function to cast numeric columns to a different numeric type (e.g. DoubleType to IntegerType).

val priceTSRdd = ... // A TimeSeriesRDD with schema Schema("time" -> LongType, "id", "price" -> DoubleType)
val result = priceTSRdd.cast("price" -> IntegerType)

• keepRows (or deleteRows) A function to filter rows based on given criteria.

val priceTSRdd = ... // A TimeSeriesRDD with columns "time", "id", and "price"
val result = priceTSRdd.keepRows { row: Row => row.getAs[Double]("price") > 100.0 }

• deleteColumns (or keepColumns) A function to filter columns by names.

val priceTSRdd = ... // A TimeSeriesRDD with columns "time", "id", and "price"
val result1 = priceTSRdd.keepColumns("time") // A TimeSeriesRDD with only "time" column
val result2 = priceTSRdd.deleteColumns("id") // A TimeSeriesRDD with only "time" and "price" columns

• renameColumns A function to modify column names without changing corresponding data types, e.g.

val priceTSRdd = ... // A TimeSeriesRDD with columns "time", "id", and "price"
val result = priceTSRdd.renameColumns("id" -> "ticker", "price" -> "highPrice")

• setTime A function to modify the time column, e.g.

val priceTSRdd = ... // A TimeSeriesRDD with columns "time", "id", and "price"
val result = priceTSRdd.setTime {
  row: Row =>
    // Set the new time to the closest trading time to the current time.
    nextClosestTradingTime(row.get("id"), row.getAs[Long]("time"))
}

• addColumns A function to add to a row with one or more new columns whose values are calculated by using only values from a row. For example, we have a TimeSeriesRDD with three columns "time", "highPrice", and "lowPrice", and we want to add a new column named "diff" to calculte the difference of the "highPrice" and "lowPrice".

val priceTSRdd = ... // A TimeSeriesRDD with columns "time", "highPrice", and "lowPrice"
val results = priceTSRdd.addColumns(
  "diff" -> DoubleType -> {
    r: Row => r.getAs[Double]("highPrice") - r.getAs[Double]("lowPrice")
  }
)
// A TimeSeriesRDD with a new column "diff" = "highPrice" - "lowPrice"

• addColumnsForCycle A cycle is defined as a list of rows that share exactly the same timestamps. For a column to be added and a list of rows in a cycle, one could use this function to add the same or different values for each row in that cycle.

val priceTSRdd = ...
// A TimeSeriesRDD with columns "time", "id", and "sellingPrice"
// time  id  sellingPrice
// ----------------------
// 1000L 0   1.0
// 1000L 1   2.0
// 1000L 1   3.0
// 2000L 0   3.0
// 2000L 0   4.0
// 2000L 1   5.0
// 2000L 2   6.0

val results = priceTSRdd.addColumnsForCycle(
  "adjustedSellingPrice" -> DoubleType -> { rows: Seq[Row] =>
    rows.map { row => (row, row.getAs[Double]("sellingPrice") * rows.size) }.toMap
  }
)
// time  id  sellingPrice adjustedSellingPrice
// -------------------------------------------
// 1000L 0   1.0          3.0
// 1000L 1   2.0          6.0
// 1000L 1   3.0          9.0
// 2000L 0   3.0          12.0
// 2000L 0   4.0          16.0
// 2000L 1   5.0          20.0
// 2000L 2   6.0          24.0

A group function is to group rows with nearby (or exactly the same) timestamps.

• groupByCycle A function to group rows within a cycle, i.e. rows with exactly the same timestamps. For example,

val priceTSRdd = ...
// A TimeSeriesRDD with columns "time" and "price"
// time  price
// -----------
// 1000L 1.0
// 1000L 2.0
// 2000L 3.0
// 2000L 4.0
// 2000L 5.0

val results = priceTSRdd.groupByCycle()
// time  rows
// ------------------------------------------------
// 1000L [[1000L, 1.0], [1000L, 2.0]]
// 2000L [[2000L, 3.0], [2000L, 4.0], [2000L, 5.0]]

• groupByInterval A funcion to group rows whose timestamps falling into an interval. Intervals could be defined by another TimeSeriesRDD. Its timestamps will be used to defined intervals, i.e. two sequential timestamps define an interval. For example,

val priceTSRdd = ...
// A TimeSeriesRDD with columns "time" and "price"
// time  price
// -----------
// 1000L 1.0
// 1500L 2.0
// 2000L 3.0
// 2500L 4.0

val clockTSRdd = ...
// A TimeSeriesRDD with only column "time"
// time
// -----
// 1000L
// 2000L
// 3000L

val results = priceTSRdd.groupByInterval(clockTSRdd)
// time  rows
// ----------------------------------
// 1000L [[1000L, 1.0], [1500L, 2.0]]
// 2000L [[2000L, 3.0], [2500L, 4.0]]

• addWindows For each row, this function adds a new column whose value for a row is a list of rows within its window.

val priceTSRdd = ...
// A TimeSeriesRDD with columns "time" and "price"
// time  price
// -----------
// 1000L 1.0
// 1500L 2.0
// 2000L 3.0
// 2500L 4.0

val result = priceTSRdd.addWindows(Window.pastAbsoluteTime("1000ns"))
// time  price window_past_1000ns
// ------------------------------------------------------
// 1000L 1.0   [[1000L, 1.0]]
// 1500L 2.0   [[1000L, 1.0], [1500L, 2.0]]
// 2000L 3.0   [[1000L, 1.0], [1500L, 2.0], [2000L, 3.0]]
// 2500L 4.0   [[1500L, 2.0], [2000L, 3.0], [2500L, 4.0]]

A temporal join function is a join function defined by a matching criteria over time. A tolerance in temporal join matching criteria specifies how much it should look past or look futue.

• leftJoin A function performs the temporal left-join to the right TimeSeriesRDD, i.e. left-join using inexact timestamp matches. For each row in the left, append the most recent row from the right at or before the same time. An example to join two TimeSeriesRDDs is as follows.

val leftTSRdd = ...
val rightTSRdd = ...
val result = leftTSRdd.leftJoin(rightTSRdd, tolerance = "1day")

• futureLeftJoin A function performs the temporal future left-join to the right TimeSeriesRDD, i.e. left-join using inexact timestamp matches. For each row in the left, appends the closest future row from the right at or after the same time.

val result = leftTSRdd.futureLeftJoin(rightTSRdd, tolerance = "1day")

Summarize functions are the functions to apply summarizer(s) to rows within a certain period, like cycle, interval, windows, etc.

• summarizeCycles A function computes aggregate statistics of rows that are within a cycle, i.e. rows share a timestamp.

val volTSRdd = ...
// A TimeSeriesRDD with columns "time", "id", and "volume"
// time  id volume
// ------------
// 1000L 1  100
// 1000L 2  200
// 2000L 1  300
// 2000L 2  400

val result = volTSRdd.summarizeCycles(Summary.sum("volume"))
// time  volume_sum
// ----------------
// 1000L 300
// 2000L 700

Similarly, we could summarize over intervals, windows, or the whole time series data set. See

• summarizeIntervals
• summarizeWindows
• addSummaryColumns

One could check timeseries.summarize.summarizer for different kinds of summarizer(s), like ZScoreSummarizer, CorrelationSummarizer, NthCentralMomentSummarizer etc.

flint.math.stat.regression aims to provide a library similar to apache statistics package and python statsmodels package.

import breeze.linalg.DenseVector
import com.twosigma.flint.math.stat.regression._

// Generate a random data set from a linear model with beta = [1.0, 2.0] and intercept = 3.0
val data = WeightedLabeledPoint.generateSampleData(sc, DenseVector(1.0, 2.0), 3.0)

// Fit the data using the OLS linear regression.
val model = OLSMultipleLinearRegression.regression(data)

// Retrieve the estimate beta and intercept.
model.estimateRegressionParameters

The followiing table list different implementations cross different packages or libraries.

• flint.math.stat - flint.math.stat.regression.LinearRegressionModel
• apache.commons.math3 - apache.commons.math3.stat.regression.OLSMultipleLinearRegression
• statsmodels - statsmodels.api in Python

In order to accept your code contributions, please fill out the appropriate Contributor License Agreement in the cla folder and submit it to [email protected].

Apache Spark is a trademark of The Apache Software Foundation. The Apache Software Foundation is not affiliated, endorsed, connected, sponsored or otherwise associated in any way to Two Sigma, Flint, or this website in any manner.

© Two Sigma Open Source, LLC