基于Spark平台的GPU调度研究整理版.pptVIP

* * KMeans Kernel耗时 为什么SparkGPUKMeans会比OpenCLKMeans多消耗那么多时间？ 操作 耗时(ms) kernel 32514 transform 117 zip 703 map和reduceByKey 610 collectAsMap 29378 /*Return the key-value pairs in this RDD to the master as a Map*/ def collectAsMap(): Map[K, V] = { val data = self.collect() val map = new mutable.HashMap[K, V] map.sizeHint(data.length) data.foreach { pair = map.put(pair._1, pair._2) } map } def collect(): Array[T] = { val results = sc.runJob(this, (iter: Iterator[T]) = iter.toArray) Array.concat(results: _*) } rdd.collect() RDD Objects