ODIn/spark-instrumented-optimizer
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity

[SPARK-19759][ML] not using blas in ALSModel.predict for optimization

Browse source 
## What changes were proposed in this pull request?

In `ALS.predict` currently we are using `blas.sdot` function to perform a dot product on two `Seq`s. It turns out that this is not the most efficient way.

I used the following code to compare the implementations:

```
def time[R](block: => R): Unit = {
    val t0 = System.nanoTime()
    block
    val t1 = System.nanoTime()
    println("Elapsed time: " + (t1 - t0) + "ns")
}
val r = new scala.util.Random(100)
val input = (1 to 500000).map(_ => (1 to 100).map(_ => r.nextFloat).toSeq)
def f(a:Seq[Float], b:Seq[Float]): Float = {
    var r = 0.0f
    for(i <- 0 until a.length) {
        r+=a(i)*b(i)
    }
    r
}
import com.github.fommil.netlib.BLAS.{getInstance => blas}
val b = (1 to 100).map(_ => r.nextFloat).toSeq
time { input.foreach(a=>blas.sdot(100, a.toArray, 1, b.toArray, 1)) }
// on average it takes 2968718815 ns
time { input.foreach(a=>f(a,b)) }
// on average it takes 515510185 ns
```

Thus this PR proposes the old-style for loop implementation for performance reasons.

## How was this patch tested?

existing UTs

Author: Marco Gaido <mgaido@hortonworks.com>

Closes #19685 from mgaido91/SPARK-19759.
This commit is contained in:
Marco Gaido 2017-11-11 04:10:54 -06:00 committed by Sean Owen
parent 808e886b96
commit 3eb315d714
1 changed files with 7 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala
View file

@ -289,9 +289,13 @@ class ALSModel private[ml] (
private val predict = udf { (featuresA: Seq[Float], featuresB: Seq[Float]) =>
if (featuresA != null && featuresB != null) {
// TODO(SPARK-19759): try dot-producting on Seqs or another non-converted type for
// potential optimization.
blas.sdot(rank, featuresA.toArray, 1, featuresB.toArray, 1)
var dotProduct = 0.0f
var i = 0
while (i < rank) {
dotProduct += featuresA(i) * featuresB(i)
i += 1
}
dotProduct
} else {
Float.NaN
}