Transcript ppt
大规模数据处理/云计算 Lecture 2 – Mapreduce System 彭波 北京大学信息科学技术学院 4/22/2011 http://net.pku.edu.cn/~course/cs402/ Jimmy Lin University of Maryland 课程建设 SEWMGroup This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States See http://creativecommons.org/licenses/by-nc-sa/3.0/us/ for details Outline • 2003 "The Google file system," in sosps. Bolton Landing, NY, USA: ACM Press, 2003. • 2004 "MapReduce: Simplified Data Processing on Large Clusters," in Osdi, 2004, 2 MapReduce Basic Typical Large-Data Problem Iterate over a large number of records Extract something of interest from each Shuffle and sort intermediate results Aggregate intermediate results Generate final output Key idea: provide a functional abstraction for these two operations 4 (Dean and Ghemawat, OSDI 2004) MapReduce Programmers specify two functions: map (k, v) → <k’, v’>* reduce (k’, v’) → <k’ ’, v’ ’>* All values with the same key are sent to the same reducer The execution framework handles everything else… 5 k1 v1 k2 v2 map a 1 k3 v3 k4 v4 map b 2 c 3 k5 v5 k6 v6 map c 6 a 5 map c 2 b 7 c 8 Shuffle and Sort: aggregate values by keys a 1 5 b 2 7 c 2 3 6 8 reduce reduce reduce r1 s1 r2 s2 r3 s3 6 MapReduce Programmers specify two functions: map (k, v) → <k’, v’>* reduce (k’, v’) → <k’, v’>* All values with the same key are sent to the same reducer The execution framework handles everything else… What’s “everything else”? 7 MapReduce “Runtime” Handles scheduling Handles “data distribution” Gathers, sorts, and shuffles intermediate data Handles errors and faults Moves processes to data Handles synchronization Assigns workers to map and reduce tasks Detects worker failures and restarts Everything happens on top of a distributed FS (later) 8 MapReduce Programmers specify two functions: map (k, v) → <k’, v’>* reduce (k’, v’) → <k’, v’>* All values with the same key are reduced together The execution framework handles everything else… Not quite…usually, programmers also specify: partition (k’, number of partitions) → partition for k’ Often a simple hash of the key, e.g., hash(k’) mod n Divides up key space for parallel reduce operations combine (k’, v’) → <k’, v’>* Mini-reducers that run in memory after the map phase Used as an optimization to reduce network traffic 9 k1 v1 k2 v2 map a 1 k4 v4 map b 2 c 3 combine a 1 k3 v3 c 6 a 5 map c 2 b 7 combine c 9 partition k6 v6 map combine b 2 k5 v5 a 5 partition c 8 combine c 2 b 7 partition c 8 partition Shuffle and Sort: aggregate values by keys a 1 5 b 2 7 c 2 3 9 6 8 8 reduce reduce reduce r1 s1 r2 s2 r3 s3 10 Two more details… Barrier between map and reduce phases But we can begin copying intermediate data earlier Keys arrive at each reducer in sorted order No enforced ordering across reducers 11 “Hello World”: Word Count Map(String docid, String text): for each word w in text: Emit(w, 1); Reduce(String term, Iterator<Int> values): int sum = 0; for each v in values: sum += v; Emit(term, value); 12 MapReduce can refer to… The programming model The execution framework (aka “runtime”) The specific implementation Usage is usually clear from context! 13 MapReduce Implementations Google has a proprietary implementation in C++ Hadoop is an open-source implementation in Java Bindings in Java, Python An Apache project Large contribution of development led by Yahoo, used in production Rapidly expanding software ecosystem Lots of custom research implementations For GPUs, cell processors, etc. 14 User Program (1) submit Master (2) schedule map (2) schedule reduce worker split 0 split 1 split 2 split 3 (5) remote read (3) read worker worker (6) write output file 0 (4) local write split 4 worker output file 1 worker Input files Map phase Intermediate files (on local disk) Reduce phase Output files 15 Adapted from (Dean and Ghemawat, OSDI 2004) How do we get data to the workers? NAS SAN Compute Nodes What’s the problem here? 16 Distributed File System Don’t move data to workers… move workers to the data! Why? Store data on the local disks of nodes in the cluster Start up the workers on the node that has the data local Not enough RAM to hold all the data in memory Disk access is slow, but disk throughput is reasonable A distributed file system is the answer GFS (Google File System) for Google’s MapReduce HDFS (Hadoop Distributed File System) for Hadoop 17 GFS: Assumptions Commodity hardware over “exotic” hardware High component failure rates Multi-gigabyte files are common, if not encouraged Files are write-once, mostly appended to Inexpensive commodity components fail all the time “Modest” number of huge files Scale “out”, not “up” Perhaps concurrently Large streaming reads over random access High sustained throughput over low latency 18 GFS slides adapted from material by (Ghemawat et al., SOSP 2003) GFS: Design Decisions Files stored as chunks Reliability through replication Simple centralized management No data caching Each chunk replicated across 3+ chunkservers Single master to coordinate access, keep metadata Fixed size (64MB) Little benefit due to large datasets, streaming reads Simplify the API Push some of the issues onto the client (e.g., data layout) HDFS = GFS clone (same basic ideas) 19 From GFS to HDFS Terminology differences: GFS master = Hadoop namenode GFS chunkservers = Hadoop datanodes Functional differences: No file appends in HDFS (planned feature) HDFS performance is (likely) slower For the most part, we’ll use the Hadoop terminology… 20 HDFS Architecture HDFS namenode Application (file name, block id) HDFS Client /foo/bar File namespace block 3df2 (block id, block location) instructions to datanode (block id, byte range) block data datanode state HDFS datanode HDFS datanode Linux file system Linux file system … … 21 Adapted from (Ghemawat et al., SOSP 2003) Namenode Responsibilities Managing the file system namespace: Coordinating file operations: Holds file/directory structure, metadata, file-to-block mapping, access permissions, etc. Directs clients to datanodes for reads and writes No data is moved through the namenode Maintaining overall health: Periodic communication with the datanodes Block re-replication and rebalancing Garbage collection 22 Putting everything together… namenode job submission node namenode daemon jobtracker tasktracker tasktracker tasktracker datanode daemon datanode daemon datanode daemon Linux file system Linux file system Linux file system … slave node … slave node … slave node 23 References 2003 "The Google file system," in sosps. Bolton Landing, NY, USA: ACM Press, 2003. 2004 "MapReduce: Simplified Data Processing on Large Clusters," in Osdi, 2004, 24 Q&A? Thanks you!