Github user MLnick commented on a diff in the pull request:
https://github.com/apache/spark/pull/17742#discussion_r114397947
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/recommendation/MatrixFactorizationModel.scala
---
@@ -276,44 +277,53 @@ object MatrixFactorizationModel extends
Loader[MatrixFactorizationModel] {
num: Int): RDD[(Int, Array[(Int, Double)])] = {
val srcBlocks = blockify(rank, srcFeatures)
val dstBlocks = blockify(rank, dstFeatures)
- val ratings = srcBlocks.cartesian(dstBlocks).flatMap {
- case ((srcIds, srcFactors), (dstIds, dstFactors)) =>
- val m = srcIds.length
- val n = dstIds.length
- val ratings = srcFactors.transpose.multiply(dstFactors)
- val output = new Array[(Int, (Int, Double))](m * n)
- var k = 0
- ratings.foreachActive { (i, j, r) =>
- output(k) = (srcIds(i), (dstIds(j), r))
- k += 1
+ /**
+ * Use dot to replace blas 3 gemm is the key approach to improve
efficiency.
+ * By this change, we can get the topK elements of each block to
reduce the GC time.
+ * Comparing with BLAS.dot, hand-written dot is high efficiency.
+ */
+ val ratings = srcBlocks.cartesian(dstBlocks).flatMap { case (srcIter,
dstIter) =>
+ val m = srcIter.size
+ val n = math.min(dstIter.size, num)
+ val output = new Array[(Int, (Int, Double))](m * n)
+ var j = 0
+ srcIter.foreach { case (srcId, srcFactor) =>
+ val pq = new BoundedPriorityQueue[(Int,
Double)](n)(Ordering.by(_._2))
--- End diff --
Here we are creating O(blocksize) PQ objects. But we could have only one,
and reuse it. We could create it outside the `foreach` calls and call `clear`
after using it. Not really sure if it will make a large difference, but could
further reduce GC
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
