1 / 18

Improving MapReduce Performance in Heterogeneous Environments

Improving MapReduce Performance in Heterogeneous Environments. Matei Zaharia, Andy Konwinski, Anthony Joseph, Randy Katz, Ion Stoica University of California at Berkeley. Motivation. MapReduce programming model growing in popularity

doria
Download Presentation

Improving MapReduce Performance in Heterogeneous Environments

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Improving MapReduce Performance in Heterogeneous Environments Matei Zaharia, Andy Konwinski,Anthony Joseph, Randy Katz, Ion Stoica University of California at Berkeley

  2. Motivation • MapReduce programming model growing in popularity • Open-source implementation, Hadoop, used at Yahoo, Facebook, CMU, Berkeley,… • Virtualized computing services like Amazon EC2 provide on-demand compute power, but less control over performance

  3. Results • Main challenge for Hadoop on EC2 was node performance heterogeneity • Designed heterogeneity-aware scheduler that improves response time up to 2x

  4. Outline • MapReduce background • The challenge of heterogeneity • LATE: a heterogeneity-aware scheduler

  5. What is MapReduce? • Programming model to split computations into independent parallel tasks • Map tasks filter data set • Reduce tasks aggregate values by key • Goal: hide the complexity of distributed programming and fault tolerance

  6. Fault Tolerance in MapReduce • Nodes fail  re-run tasks • Nodes very slow (stragglers)  launch backup copies of tasks • How to do this in heterogeneous env.?

  7. Heterogeneity in Virtualized Environments • VM technology isolates CPU and memory, but disk and network are shared • Full bandwidth when no contention • Equal shares when there is contention • 2.5x I/O performance difference on EC2

  8. Disk Performance Heterogeneity Experiment

  9. Backup Task Scheduling in Hadoop • Scheduler starts all primary tasks, then looks for tasks to back up • Tasks report “progress score” from 0 to 1 • Backup launched if progress < avgProgress - 20%

  10. Problems in Heterogeneous Environment • Too many backups (trash shared resources) • Wrong tasks may be backed up • Backups may be placed on slow nodes • Tasks never backed up if progress > 80% • Result: 80% of reduces backed up in some experiments, network overloaded

  11. Progress Rate Approaches • Compute average progress rate, back up tasks that are “far enough” below this • Problems: • How long to wait for statistics? • Can still select the wrong tasks

  12. Example Node 1 1 task/min Node 2 3x slower 1.9x slower Node 3 Time (min)

  13. Example What if the job had 5 tasks? 2 min Node 1 Node 2 time left: 1 min Node 3 time left: 1.8 min Time (min) Should back up node 3’s task

  14. Our Scheduler: LATE • Insight: back up the task with the largest estimated finish time • “Longest Approximate Time to End” • Sanity thresholds: • Cap backup tasks to ~10% • Launch backups on fast nodes • Only back up tasks that are sufficiently slow

  15. LATE Details • Estimating finish time: • Thresholds: • 25th percentiles for slow node/task, 10% cap • Sensitivity analysis shows robustness progress score progress rate = execution time 1 – progress score estimated time left = progress rate

  16. Evaluation • EC2 experiments (3 job types, 200 nodes) • Experiments in small controlled testbed • Contention through VM placement • Results: • 2x better response time if there are stragglers • 30% better response time when no stragglers

  17. Conclusion • Heterogeneity is a challenge for parallel applications, and is getting more important • Lessons for scheduling backup tasks: • Detecting slow nodes isn’t enough; do it early • Pick tasks which hurt response time the most • Be mindful of shared resources • 2x improvement using simple algorithm

  18. Questions? ? ? ?

More Related