Data set cloudrank-d-hpca_tutorial
- 1. INSTITUTEOFCOMPUTINGTECHNOLOGY CloudRank-D：A Benchmark Suite for Private Cloud Systems Jing Quan Institute of Computing Technology, Chinese Academy of Sciences and University of Science and Technology of China 1 HVC tutorial in conjunction with The 19th IEEE International Symposium on High Performance Computer Architecture (HPCA 2013)
- 2. HVC TutorialHPCA 2013 Contents • Background & Motivation • Introduction of CloudRank-D • Use cases
- 3. HVC TutorialHPCA 2013 Contents • Background & Motivation • Introduction of CloudRank-D • Use cases
- 4. HVC TutorialHPCA 2013 What is Private Cloud ? • Private Cloud – The cloud infrastructure is provisioned for exclusive use by a single organization comprising multiple consumers (e.g., business units). It may be owned, managed, and operated by the organization, a third party, or some combination of them, and it may exist on or off premises. "The NIST Definition of Cloud Computing" National Institute of Standards and Technology. Retrieved 24 July 2011 http://blogs.technet.com/b/yungchou/archive/2011/03/21/what-is- private-cloud.aspx
- 5. HVC TutorialHPCA 2013 Typical Data Processing Application Recommender systerm Social Network …… Search Engine Hadoop Master Node Job Production Client Front End MapReduce Jobs Job Deployment Scheduler Job flow Framework Node Node …… Node HDFS
- 6. HVC TutorialHPCA 2013 User Concerns Xeon Xeon Xeon Xeon Atom Atom Atom Atom How to quantitatively measure systems? Which one is better (ranking systems)? How to guide optimization?
- 7. HVC TutorialHPCA 2013 What is CloudRank-D? CloudRank-D Private cloud systems Ranking systems Data processing General Description CloudRank-D is a benchmark suite, used to evaluate private cloud systems that is shared for running data processing applications.
- 8. HVC TutorialHPCA 2013 Why CloudRank-D? Benchmark Target of Evaluation MineBench Data mining algorithms GridMix Hadoop framework HiBench Hadoop framework WL suite Hadoop framework CloudRank-D The whole system
- 9. HVC TutorialHPCA 2013 Our Focus: Evaluating the Whole System Applications (Data analysis) Framework (Hadoop) System platform System platform Default framework (Hadoop) Applications (Data analysis) Performanc of Software & Hardware CloudRank-DGridMix etc. Hadoop Performance vs
- 10. HVC TutorialHPCA 2013 Comparison of Different Benchmarks Suites Mine- Bench Grid- Mix HiBench WL suite CloudSuite CloudRank-D Representa- tive applications Basic operations n y y y n y Classification y n y n y y Clustering y n y n n y Recommend- ation n n n n n y Sequence learning y n n n n y Association rule mining y n n n n y Data warehouse operations n n n y n y
- 11. HVC TutorialHPCA 2013 Comparison of Different Benchmarks Suites(Cont') MineB ench Grid Mix HiBench WL suite CloudSuite CloudRank-D Workloads description Submission pattern n n n y n y Scheduling strategies n n n n n y System software configuration n n n n n y Data models n n n n n y Data semantics n n n n n y Scalable data size y y n y n y Category of datacentric computation n n n y n y
- 12. HVC TutorialHPCA 2013 Contents • Background & Motivation • Introduction of CloudRank-D – Methodology • Use cases
- 13. HVC TutorialHPCA 2013 CloudRank-D Methodology System platform Workloads with usage patterns Performance reports running feedback Get the peak system performance Ⅰ.Measure systems Ⅱ.Find a suitable system Ⅲ.Optimize systems
- 14. HVC TutorialHPCA 2013 Configurable Workloads with Tunable Usage Patterns Scalable applications and input datasets Tunable submission patterns Configurable runtime system • Representive applications domains • User specific • Scalable data size • Modeling production system logs • Experiences from industry and academic
- 15. HVC TutorialHPCA 2013 Usage patterns Scalable applications and input data sets Tunable submission patterns Configurable framework CloudRank-D Methodology: Workloads with Usage Patterns
- 16. HVC TutorialHPCA 2013 Scalable Applications and Input Data Sets Scalable applications and input data sets Submitted jobs composed of appropriate applications Expanded data sets
- 17. HVC TutorialHPCA 2013 NO. Category Application Data size Data semantics 1 basic operation sort scalable (scale to 10PB) automatically generated 2 word count 3 grep 4 classification naive bayes 5 support vector machine Scientist Search 6 cluster k-means scalable sougou corpus 7 recommenda tion Item based collaborative filtering scalable ratings on movies Applications and Input Data Sets
- 18. HVC TutorialHPCA 2013 NO. Category Application Data size Data semantics 8 association rule mining frequent pattern growth fixed retail market basket data click-stream data , traffic accident data, collection of web html documents 9 sequence learning hidden morkov model scalable Scientist Search 10 warehouse operation grep select automatically generated table 11 ranking select 12 aggregation 13 uservisits-ranking join Applications and Input Data Sets (Cont') You can add any applications you want !
- 19. HVC TutorialHPCA 2013 Applications Combinations Demonstration WebCrawling DataMining MachineLearning ImageProcessing Naive Bayes & SVM HMM & IBCF & FPG 35% TextIndexing LogProcessing Basic Operations 31% Reporting DataStorage Hive 34% wiki.apache.org/hadoop/PoweredBy
- 20. HVC TutorialHPCA 2013 Data Set Sizes Demonstration Map Number Percentage Size <10 40.57% 128MB~1.25GB 10~500 39.33% 1.25GB~62.5GB 500~2000 12.03% 63.5GB~250GB >2000 8.07% 250GB~ Workload Characterization on a Production Hadoop Cluster: A Case Study on Taobao
- 21. HVC TutorialHPCA 2013 Usage patterns Scalable applications and input data size Tunable submission patterns Configurable framework Workloads with usage patterns
- 22. HVC TutorialHPCA 2013 Submission Patterns Submission patterns Submission intervals Submission orders
- 23. HVC TutorialHPCA 2013 Submission Intervals Form the Facebook report, distribution of inter-arrival times was roughly exponential with a mean of 14 seconds. Ddelay scheduling: A simple technique for achieving locality and fairness in cluster scheduling. In Proceeding In Proceedings of the 5th European conference on Computer systems. Probability density function
- 24. HVC TutorialHPCA 2013 Submission Orders • For the workloads with different resource sizes and different catelogs – Submitting jobs randomly – Submitting jobs with batch model
- 25. HVC TutorialHPCA 2013 Usage patterns Scalable applications and input data size Tunable submission patterns Configurable framework Workloads with usage patterns
- 26. HVC TutorialHPCA 2013 Hadoop Configurations Dimensions Explanation Map/Reduce Number affect system utilization Scheduling Policy Hadoop chooses which job to run according to this policy Main Parameters mapred.tasktracker.map.tasks.maximum mapred.tasktracker.reduce.tasks.maxmum mapred.child.java.opts dfs.block.size
- 27. HVC TutorialHPCA 2013 Hadoop Settings Parameter Value Mapred.tasktracker.tasks.r educe.maximum usually, this value is equal to the core number of current node dis.block.size default value is 64M, you can change it to ensure there won't be too much map number for most workloads Map (adjust through the block size) 10~100 per node, and it's would be better if the execution time was more than 1min
- 28. HVC TutorialHPCA 2013 Scheduling Policy • Common schedule algorithms – First input first out – Fair-share scheduler – Capacity scheduler • Fair-share scheduling can do a good job Workload Characterization on a Production Hadoop Cluster: A Case Study on Taobao
- 29. HVC TutorialHPCA 2013 CloudRank-D methodology: Our metrics • Focus – From user perspective – Easy to compare and understand • Metrics – Data processed per second or joule • How to get it? DPS=Total data input size/Total run time DPJ=Total data input size/Total energy consumption
- 30. HVC TutorialHPCA 2013 Contents • Background & Motivation • Introduction of CloudRank-D • Use cases
- 31. HVC TutorialHPCA 2013 How to use? CloudRank-D
- 32. HVC TutorialHPCA 2013 Use Case 1: Comparing Two Hardware Platforms Cluster 1 Cluster 2 Xeon Xeon Xeon Xeon Atom Atom Atom Atom Two clusters comprise 128 nodes respectively.
- 33. HVC TutorialHPCA 2013 Procedures Step 1 Prepared hardware platform Step 1 Build foundation platform Step 2 Customize workloads Step 3 Run workloads Step 4 Get results and optimize systems
- 34. HVC TutorialHPCA 2013 Base Information • Evaluating two private cloud systems • Using all workloads we provide • Deploying uniform software platform • Adopting same configuration
- 35. HVC TutorialHPCA 2013 Software Configuration software stack Hadoop version 0.20.2 Hive version 0.6.0 Mahout version 0.6 map/reduce slot 4 map slots and 2 reduce slots Hadoop system configuration default Hadoop scheduling algorithm fair schedule
- 36. HVC TutorialHPCA 2013 Run your workloads Job Submission Patterns You can submit the workloads according to the exponential distribution with a specified mean submission interval --- 14 seconds Submission order : Random
- 37. HVC TutorialHPCA 2013 An example of result The comparion between Xeon Atom on two metrics • Xeon – less time, more energy • Atom – more time, less energy AtomXeon Totaldataprocessed persecond(KB/S) 4000 2000 0 Xeon Atom 10 0 Totaldataprocessed perjoule(KB/J) 5
- 38. HVC TutorialHPCA 2013 Optimized (Cont') • Tuning the interval We can see that the best performance occurred when the interval value is 70.
- 39. HVC TutorialHPCA 2013 Use Case 2: Scheduling Evaluation 39/ I have designed a new Hadoop scheduling algorithm, but I don’t have the workloads for test. How to evaluate the scheduling ? And let people trust the evaluations results.
- 40. HVC TutorialHPCA 2013 Using CloudRank-D Step 1 Building foundation platform with different scheduling policy Step 1 Build foundation platform Step 2 Customizing workloads with productive scenarios Step 3 Running workloads Step 4 Getting the metrics under different scheduling policy
- 41. HVC TutorialHPCA 2013 Our Result 5 0 Totaldataprocessedperjoule(KB/J) Fair scheduler FIFO scheduler 4000 2000 0 Totaldataprocessedpersecond(KB/S) Fair scheduler FIFO scheduler We can see that fair scheduler works better than FIFO scheduler.
- 42. HVC TutorialHPCA 2013 • Contact us – Websit: http://prof.ict.ac.cn/CloudRank/ – Email: quanjing@ict.ac.cn