Scala优化数据框

Question

如何在存储，数据移动和处理中利用优化的Scala Spark Dataframe？

  val lasVegasBusiness = business.filter("city=='Las Vegas'")
  val stars = review.withColumn("stars",col("stars").cast(DoubleType))
  val startReview = review.filter("stars > 4")
  lasVegasBusiness.join(startReview,lasVegasBusiness("business_id") === 
  startReview("business_id"),"inner")