[SPARK-11560] Optimize KMeans implementation / remove 'runs' from implementation

• Type: Improvement
• Status: Resolved
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: 1.6.0
• Fix Version/s: 2.1.0
• Component/s: MLlib
• Assignee: Sean R. Owen
• Reporter: Xiangrui Meng
• Shepherd: Xiangrui Meng
• Created: 06/Nov/15 19:45
• Updated: 12/Oct/16 09:01
• Resolved: 12/Oct/16 09:01

Description

After we dropped `runs`, we can simplify and optimize the k-means implementation.

Issue Links

• is blocked by: SPARK-11559 Make `runs` no effect in k-means
• is related to: SPARK-12450 Un-persist broadcasted variables in KMeans
• is related to: SPARK-17389 KMeans speedup with better choice of k-means|| init steps = 2
• relates to: SPARK-8519 Blockify distance computation in k-means
• links to: [Github] Pull Request #10306 (yanboliang)
• links to: [Github] Pull Request #10806 (yanboliang)
• links to: [Github] Pull Request #14937 (yanboliang)
• links to: [Github] Pull Request #15342 (srowen)