Cloud Computing and MapReduce Used slides from RAD Lab at UC Berkeley about the cloud ( and slides from Jimmy Lin’s slides ( (licensed under Creation Commons Attribution 3.0 License)

Cloud computing What is the “cloud”? –Many answers. Easier to explain with examples: Gmail is in the cloud Amazon (AWS) EC2 and S3 are the cloud Google AppEngine is the cloud Windows Azure is the cloud SimpleDB is in the cloud The “network” (cloud) is the computer

Cloud Computing What about Wikipedia? “Cloud computing is the delivery of computing as a service rather than a product, whereby shared resources, software, and information are provided to computers and other devices as a utility (like the electricity grid) over a network (typically the Internet). “ computingservice productutilityelectricity gridnetworkInternet

Cloud properties Cloud offers: –Scalability : scale out vs scale up (also scale back) –Reliability (hopefully!) –Availability (24x7) –Elasticity : pay-as-you go depending on your demand –Multi-tenancy

more Scalability means that you (can) have infinite resources, can handle unlimited number of users Multi-tenancy enables sharing of resources and costs across a large pool of users. Lower cost, higher utilization… but other issues: e.g. security. Elasticity: you can add or remove computer nodes and the end user will not be affected/see the improvement quickly. Utility computing (similar to electrical grid)

CLOUD COMPUTING ECONOMICS AND ELASTICITY

Cloud Application Demand Many cloud applications have cyclical demand curves –Daily, weekly, monthly, … Workload spikes more frequent and significant –Death of Michael Jackson: 22% of tweets, 20% of Wikipedia traffic, Google thought they are under attack –Obama inauguration day: 5x increase in tweets Demand Time Resources

Unused resources Economics of Cloud Users Pay by use instead of provisioning for peak Recall: DC costs >$150M and takes 24+ months to design and build Static data centerData center in the cloud Demand Capacity Time Resources Demand Capacity Time Resources How do you pick a capacity level?

Unused resources Economics of Cloud Users Risk of over-provisioning: underutilization Huge sunk cost in infrastructure Static data center Demand Capacity Time Resources Demand Capacity Time (days) 1 23

Utility Computing Arrives Amazon Elastic Compute Cloud (EC2) “Compute unit” rental: $ /hour –1 CU ≈ GHz 2007 AMD Opteron/Intel Xeon core No up-front cost, no contract, no minimum Billing rounded to nearest hour (also regional,spot pricing) New paradigm(!) for deploying services?, HPC? Northern VA cluster

Utility Storage Arrives Amazon S3 and Elastic Block Storage offer low-cost, contract-less storage

Cloud Computing Infrastructure Computation model: MapReduce* Storage model: HDFS* Other computation models: HPC/Grid Computing Network structure *Some material adapted from slides by Jimmy Lin, Christophe Bisciglia, Aaron Kimball, & Sierra Michels-Slettvet, Google Distributed Computing Seminar, 2007 (licensed under Creation Commons Attribution 3.0 License)

Cloud Computing Computation Models Finding the right level of abstraction –von Neumann architecture vs cloud environment Hide system-level details from the developers –No more race conditions, lock contention, etc. Separating the what from how –Developer specifies the computation that needs to be performed –Execution framework (“runtime”) handles actual execution

“Big Ideas” Scale “out”, not “up” –Limits of SMP and large shared-memory machines Idempotent operations –Simplifies redo in the presence of failures Move processing to the data –Cluster has limited bandwidth Process data sequentially, avoid random access –Seeks are expensive, disk throughput is reasonable Seamless scalability for ordinary programmers –From the mythical man-month to the tradable machine-hour

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 – MapReduce Map Reduce (Dean and Ghemawat, OSDI 2004)

ggggg fffff Map Fold Roots in Functional Programming

MapReduce Programmers specify two functions: map (k, v) → * reduce (k’, v’) → * –All values with the same key are sent to the same reducer The execution framework handles everything else…

map Shuffle and Sort: aggregate values by keys reduce k1k1 k2k2 k3k3 k4k4 k5k5 k6k6 v1v1 v2v2 v3v3 v4v4 v5v5 v6v6 ba12cc36ac52bc78 a15b27c2368 r1r1 s1s1 r2r2 s2s2 r3r3 s3s3 MapReduce

Programmers specify two functions: map (k, v) → * reduce (k’, v’) → * –All values with the same key are sent to the same reducer The execution framework handles everything else… What’s “everything else”?

MapReduce “Runtime” Handles scheduling –Assigns workers to map and reduce tasks Handles “data distribution” –Moves processes to data Handles synchronization –Gathers, sorts, and shuffles intermediate data Handles errors and faults –Detects worker failures and automatically restarts Handles speculative execution –Detects “slow” workers and re-executes work Everything happens on top of a distributed FS (later) Sounds simple, but many challenges!

MapReduce Programmers specify two functions: map (k, v) → * reduce (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 R –Divides up key space for parallel reduce operations combine (k’, v’) → * –Mini-reducers that run in memory after the map phase –Used as an optimization to reduce network traffic

combine ba12c9ac52bc78 partition map k1k1 k2k2 k3k3 k4k4 k5k5 k6k6 v1v1 v2v2 v3v3 v4v4 v5v5 v6v6 ba12cc36ac52bc78 Shuffle and Sort: aggregate values by keys reduce a15b27 c298 r1r1 s1s1 r2r2 s2s2 r3r3 s3s3 c2368

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

split 0 split 1 split 2 split 3 split 4 worker Master User Program output file 0 output file 1 (1) submit (2) schedule map(2) schedule reduce (3) read (4) local write (5) remote read (6) write Input files Map phase Intermediate files (on local disk) Reduce phase Output files Adapted from (Dean and Ghemawat, OSDI 2004) MapReduce Overall Architecture

“Hello World” Example: Word Count Map(String docid, String text): for each word w in text: Emit(w, 1); Reduce(String term, Iterator values): int sum = 0; for each v in values: sum += v; Emit(term, value);

MapReduce can refer to… The programming model The execution framework (aka “runtime”) The specific implementation Usage is usually clear from context!

MapReduce Implementations Google has a proprietary implementation in C++ –Bindings in Java, Python Hadoop is an open-source implementation in Java –Development led by Yahoo, used in production –Now an Apache project –Rapidly expanding software ecosystem, but still lots of room for improvement Lots of custom research implementations –For GPUs, cell processors, etc.

Cloud Computing Storage, or how do we get data to the workers? Compute Nodes NAS SAN What’s the problem here?

Distributed File System Don’t move data to workers… move workers to the data! –Store data on the local disks of nodes in the cluster –Start up the workers on the node that has the data local Why? –Network bisection bandwidth is limited –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

GFS: Assumptions Choose commodity hardware over “exotic” hardware –Scale “out”, not “up” High component failure rates –Inexpensive commodity components fail all the time “Modest” number of huge files –Multi-gigabyte files are common, if not encouraged Files are write-once, mostly appended to –Perhaps concurrently Large streaming reads over random access –High sustained throughput over low latency GFS slides adapted from material by (Ghemawat et al., SOSP 2003)

GFS: Design Decisions Files stored as chunks –Fixed size (64MB) Reliability through replication –Each chunk replicated across 3+ chunkservers Single master to coordinate access, keep metadata –Simple centralized management No data caching –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 implemented in Java)

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

Adapted from (Ghemawat et al., SOSP 2003) (file name, block id) (block id, block location) instructions to datanode datanode state (block id, byte range) block data HDFS namenode HDFS datanode Linux file system … HDFS datanode Linux file system … File namespace /foo/bar block 3df2 Application HDFS Client HDFS Architecture

Namenode Responsibilities Managing the file system namespace: –Holds file/directory structure, metadata, file-to- block mapping, access permissions, etc. Coordinating file operations: –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

Putting everything together… datanode daemon Linux file system … tasktracker slave node datanode daemon Linux file system … tasktracker slave node datanode daemon Linux file system … tasktracker slave node namenode namenode daemon job submission node jobtracker

MapReduce/GFS Summary Simple, but powerful programming model Scales to handle petabyte+ workloads –Google: six hours and two minutes to sort 1PB (10 trillion 100-byte records) on 4,000 computers –Yahoo!: hours to sort 1PB on 3,800 computers Incremental performance improvement with more nodes Seamlessly handles failures, but possibly with performance penalties