Presentation is loading. Please wait.

Presentation is loading. Please wait.

© Hortonworks Inc. 2013 Page 1 Apache Tez : Accelerating Hadoop Data Processing Bikas

Similar presentations

Presentation on theme: "© Hortonworks Inc. 2013 Page 1 Apache Tez : Accelerating Hadoop Data Processing Bikas"— Presentation transcript:

1 © Hortonworks Inc Page 1 Apache Tez : Accelerating Hadoop Data Processing Bikas

2 © Hortonworks Inc Tez – Introduction Page 2 Distributed execution framework targeted towards data- processing applications. Based on expressing a computation as a dataflow graph. Highly customizable to meet a broad spectrum of use cases. Built on top of YARN – the resource management framework for Hadoop. Open source Apache project and Apache licensed.

3 © Hortonworks Inc. 2012© Hortonworks Inc Confidential and Proprietary. Tez – Hadoop 1 ---> Hadoop 2 Monolithic Resource Management – MR Execution Engine – MR Layered Resource Management – YARN Engines – Hive, Pig, Cascading, Your App!

4 © Hortonworks Inc Tez – Empowering Applications Page 4 Tez solves hard problems of running on a distributed Hadoop environment Apps can focus on solving their domain specific problems This design is important to be a platform for a variety of applications App Tez Custom application logic Custom data format Custom data transfer technology Distributed parallel execution Negotiating resources from the Hadoop framework Fault tolerance and recovery Horizontal scalability Resource elasticity Shared library of ready-to-use components Built-in performance optimizations Security

5 © Hortonworks Inc Tez – End User Benefits Better performance of applications Built-in performance + Application define optimizations Better predictability of results Minimization of overheads and queuing delays Better utilization of compute capacity Efficient use of allocated resources Reduced load on distributed filesystem (HDFS) Reduce unnecessary replicated writes Reduced network usage Better locality and data transfer using new data patterns Higher application developer productivity Focus on application business logic rather than Hadoop internals Page 5

6 © Hortonworks Inc Tez – Design considerations Don’t solve problems that have already been solved. Or else you will have to solve them again! Leverage discrete task based compute model for elasticity, scalability and fault tolerance Leverage several man years of work in Hadoop Map-Reduce data shuffling operations Leverage proven resource sharing and multi-tenancy model for Hadoop and YARN Leverage built-in security mechanisms in Hadoop for privacy and isolation Page 6 Look to the Future with an eye on the Past

7 © Hortonworks Inc Tez – Problems that it addresses Expressing the computation Direct and elegant representation of the data processing flow Interfacing with application code and new technologies Performance Late Binding : Make decisions as late as possible using real data from at runtime Leverage the resources of the cluster efficiently Just work out of the box! Customizable to let applications tailor the job to meet their specific requirements Operation simplicity Painless to operate, experiment and upgrade Page 7

8 © Hortonworks Inc Tez – Simplifying Operations No deployments to do. No side effects. Easy and safe to try it out! Tez is a completely client side application. Simply upload to any accessible FileSystem and change local Tez configuration to point to that. Enables running different versions concurrently. Easy to test new functionality while keeping stable versions for production. Leverages YARN local resources. Page 8 Client Machine Client Machine Node Manager Node Manager TezTask Node Manager Node Manager TezTask TezClient HDFS Tez Lib 1 Tez Lib 2 Client Machine Client Machine TezClient

9 © Hortonworks Inc Tez – Expressing the computation Page 9 Aggregate Stage Partition Stage Preprocessor Stage Sampler Task-1 Task-2 Task-1 Task-2 Task-1 Task-2 Samples Ranges Distributed Sort Distributed data processing jobs typically look like DAGs (Directed Acyclic Graph). Vertices in the graph represent data transformations Edges represent data movement from producers to consumers

10 © Hortonworks Inc Tez – Expressing the computation Page 10 Tez provides the following APIs to define the processing DAG API Defines the structure of the data processing and the relationship between producers and consumers Enable definition of complex data flow pipelines using simple graph connection API’s. Tez expands the logical DAG at runtime This is how the connection of tasks in the job gets specified Runtime API Defines the interfaces using which the framework and app code interact with each other App code transforms data and moves it between tasks This is how we specify what actually executes in each task on the cluster nodes

11 © Hortonworks Inc Tez – DAG API // Define DAG DAG dag = DAG.create(); // Define Vertex Vertex Map1 = Vertex.create(Processor.class); // Define Edge Edge edge = Edge.create(Map1, Reduce1, SCATTER_GATHER, PERSISTED, SEQUENTIAL, Output.class, Input.class); // Connect them dag.addVertex(Map1).addEdge(edge)… Page 11 Defines the global processing flow Map1 Map2 Reduce1 Reduce2 Join Scatter Gather Scatter Gather

12 © Hortonworks Inc Tez – DAG API Page 12 Data movement – Defines routing of data between tasks One-To-One : Data from the i th producer task routes to the i th consumer task. Broadcast : Data from a producer task routes to all consumer tasks. Scatter-Gather : Producer tasks scatter data into shards and consumer tasks gather the data. The i th shard from all producer tasks routes to the i th consumer task. Edge properties define the connection between producer and consumer tasks in the DAG

13 © Hortonworks Inc Tez – Logical DAG expansion at Runtime Page 13 Reduce1 Map2 Reduce2 Join Map1

14 © Hortonworks Inc Tez – Runtime API Flexible Inputs-Processor-Outputs Model Thin API layer to wrap around arbitrary application code Compose inputs, processor and outputs to execute arbitrary processing Applications decide logical data format and data transfer technology Customize for performance Built-in implementations for Hadoop 2.0 data services – HDFS and YARN ShuffleService. Built on the same API. Your implementations are as first class as ours! Page 14

15 © Hortonworks Inc Tez – Task Composition Page 15 V-A V-B V-C Logical DAG Output-1 Output-3 Processor-A Input-2 Processor-B Input-4 Processor-C Task A Task B Task C Edge AB Edge AC V-A = { Processor-A.class } V-B = { Processor-B.class } V-C = { Processor-C.class } Edge AB = { V-A, V-B, Output-1.class, Input-2.class } Edge AC = { V-A, V-C, Output-3.class, Input-4.class }

16 © Hortonworks Inc Tez – Library of Inputs and Outputs Page 16 Classical ‘Map’Classical ‘Reduce’ Intermediate ‘Reduce’ for Map-Reduce-Reduce Map Processor HDFS Input Sorted Output Reduce Processor Shuffle Input HDFS Output Reduce Processor Shuffle Input Sorted Output What is built in? – Hadoop InputFormat/OutputFormat – SortedGroupedPartitioned Key-Value Input/Output – UnsortedGroupedPartitioned Key-Value Input/Output – Key-Value Input/Output

17 © Hortonworks Inc Tez – Composable Task Model Page 17 Hive Processor HDFS Input HDFS Input Remote File Server Input Remote File Server Input HDFS Output HDFS Output Local Disk Output Local Disk Output Custom Processor HDFS Input HDFS Input Remote File Server Input Remote File Server Input HDFS Output HDFS Output Local Disk Output Local Disk Output Custom Processor RDMA Input RDMA Input Native DB Input Native DB Input Kakfa Pub-Sub Output Kakfa Pub-Sub Output Amazon S3 Output Amazon S3 Output AdoptEvolveOptimize

18 © Hortonworks Inc Tez – Performance Benefits of expressing the data processing as a DAG Reducing overheads and queuing effects Gives system the global picture for better planning Efficient use of resources Re-use resources to maximize utilization Pre-launch, pre-warm and cache Locality & resource aware scheduling Support for application defined DAG modifications at runtime for optimized execution Change task concurrency Change task scheduling Change DAG edges Change DAG vertices Page 18

19 © Hortonworks Inc Tez – Benefits of DAG execution Faster Execution and Higher Predictability Eliminate replicated write barrier between successive computations. Eliminate job launch overhead of workflow jobs. Eliminate extra stage of map reads in every workflow job. Eliminate queue and resource contention suffered by workflow jobs that are started after a predecessor job completes. Better locality because the engine has the global picture Page 19 Pig/Hive - MR Pig/Hive - Tez

20 © Hortonworks Inc Tez – Container Re-Use Reuse YARN containers/JVMs to launch new tasks Reduce scheduling and launching delays Shared in-memory data across tasks JVM JIT friendly execution Page 20 YARN Container / JVM TezTask Host TezTask1 TezTask2 Shared Objects YARN Container Tez Application Master Tez Application Master Start Task Task Done Start Task

21 © Hortonworks Inc Tez – Sessions Page 21 Application Master Client Start Session Submit DAG Task Scheduler Container Pool Shared Object Registry Pre Warmed JVM Sessions Standard concepts of pre-launch and pre-warm applied Key for interactive queries Represents a connection between the user and the cluster Multiple DAGs executed in the same session Containers re-used across queries Takes care of data locality and releasing resources when idle

22 © Hortonworks Inc Tez – Re-Use in Action Task Execution Timeline Containers Time 

23 © Hortonworks Inc Tez – Customizable Core Engine Page 23 Vertex-2 Vertex-1 Start vertex Vertex Manager Start tasks DAG Scheduler DAG Scheduler Get Priority Start vertex Task Scheduler Task Scheduler Get container Vertex Manager Determines task parallelism Determines when tasks in a vertex can start. DAG Scheduler Determines priority of task Task Scheduler Allocates containers from YARN and assigns them to tasks

24 © Hortonworks Inc Tez – Event Based Control Plane Page 24 Reduce Task 2 Input1 Input2 Map Task 2 Output1 Output2 Output3 Map Task 1 Output1 Output2 Output3 AM Router Scatter-Gather Edge Events used to communicate between the tasks and between task and framework Data Movement Event used by producer task to inform the consumer task about data location, size etc. Input Error event sent by task to the engine to inform about errors in reading input. The engine then takes action by re-generating the input Other events to send task completion notification, data statistics and other control plane information Data Event Error Event

25 © Hortonworks Inc Tez – Automatic Reduce Parallelism Page 25 Map Vertex Reduce Vertex App Master Vertex Manager Data Size Statistics Vertex State Machine Vertex State Machine Set Parallelism Cancel Task Re-Route Event Model Map tasks send data statistics events to the Reduce Vertex Manager. Vertex Manager Pluggable application logic that understands the data statistics and can formulate the correct parallelism. Advises vertex controller on parallelism

26 © Hortonworks Inc Tez – Reduce Slow Start/Pre-launch Page 26 Map Vertex Reduce Vertex App Master Vertex Manager Task Completed Vertex State Machine Vertex State Machine Start Tasks Start Event Model Map completion events sent to the Reduce Vertex Manager. Vertex Manager Pluggable application logic that understands the data size. Advises the vertex controller to launch the reducers before all maps have completed so that shuffle can start.

27 © Hortonworks Inc Tez – Theory to Practice Performance Scalability Page 27

28 © Hortonworks Inc Tez – Hive TPC-DS Scale 200GB latency

29 Tez – Pig performance gains Demonstrates performance gains from a basic translation to a Tez DAG Deeper integration in the works for further improvements

30 © Hortonworks Inc Tez – Observations on Performance Number of stages in the DAG Higher the number of stages in the DAG, performance of Tez (over MR) will be better. Cluster/queue capacity More congested a queue is, the performance of Tez (over MR) will be better due to container reuse. Size of intermediate output More the size of intermediate output, the performance of Tez (over MR) will be better due to reduced HDFS usage. Size of data in the job For smaller data and more stages, the performance of Tez (over MR) will be better as percentage of launch overhead in the total time is high for smaller jobs. Offload work to the cluster Move as much work as possible to the cluster by modelling it via the job DAG. Exploit the parallelism and resources of the cluster. E.g. MR split calculation. Vertex caching The more re-computation can be avoided the better is the performance. Page 30

31 © Hortonworks Inc Tez – Data at scale Page 31 Hive TPC-DS Scale 10TB

32 © Hortonworks Inc Tez – DAG definition at scale Page 32 Hive : TPC-DS Query 64 Logical DAG

33 © Hortonworks Inc Tez – Container Reuse at Scale 78 vertices tasks on 50 containers (TPC-DS Query 4) Page 33

34 © Hortonworks Inc Tez – Real World Use Cases for the API Page 34

35 © Hortonworks Inc Tez – Broadcast Edge SELECT ss.ss_item_sk, ss.ss_quantity, avg_price, inv.inv_quantity_on_hand FROM (select avg(ss_sold_price) as avg_price, ss_item_sk, ss_quantity_sk from store_sales group by ss_item_sk) ss JOIN inventory inv ON (inv.inv_item_sk = ss.ss_item_sk); Hive – MRHive – Tez M M M MM HDFS Store Sales scan. Group by and aggregation reduce size of this input. Inventory scan and Join Broadcast edge M M M HDFS Store Sales scan. Group by and aggregation. Inventory and Store Sales (aggr.) output scan and shuffle join. R R RR R R M MMM HDFS Hive : Broadcast Join

36 © Hortonworks Inc Tez – Multiple Outputs Page 36 Pig : Split & Group-by f = LOAD ‘foo’ AS (x, y, z); g1 = GROUP f BY y; g2 = GROUP f BY z; j = JOIN g1 BY group, g2 BY group; Group by y Group by z Load foo Join Load g1 and Load g2 Group by y Group by z Load foo Join Multiple outputs Reduce follows reduce HDFS Split multiplex de-multiplex Pig – MRPig – Tez

37 © Hortonworks Inc Tez – Multiple Outputs FROM (SELECT * FROM store_sales, date_dim WHERE ss_sold_date_sk = d_date_sk and d_year = 2000) INSERT INTO TABLE t1 SELECT distinct ss_item_sk INSERT INTO TABLE t2 SELECT distinct ss_customer_sk; Hive – MRHive – Tez M M M M HDFS Map join date_dim/store sales Two MR jobs to do the distinct MMM MM HDFS RR MMM R MMM R Broadcast Join (scan date_dim, join store sales) Distinct for customer + items Materialize join on HDFS Hive : Multi-insert queries

38 © Hortonworks Inc Tez – One to One Edge Page 38 Aggregate Sample L Join Stage sample map on distributed cache l = LOAD ‘left’ AS (x, y); r = LOAD ‘right’ AS (x, z); j = JOIN l BY x, r BY x USING ‘skewed’; Load & Sample Aggregate Partition L Join Pass through input via 1-1 edge Partition R HDFS Broadcast sample map Partition L and Partition R Pig – MRPig – Tez Pig : Skewed Join

39 © Hortonworks Inc Tez – Custom Edge SELECT ss.ss_item_sk, ss.ss_quantity, inv.inv_quantity_on_hand FROM store_sales ss JOIN inventory inv ON (inv.inv_item_sk = ss.ss_item_sk); Hive – MRHive – Tez M MM M M HDFS Inventory scan (Runs on cluster potentially more than 1 mapper) Store Sales scan and Join (Custom vertex reads both inputs – no side file reads) Custom edge (routes outputs of previous stage to the correct Mappers of the next stage) M M M M HDFS Inventory scan (Runs as single local map task) Store Sales scan and Join (Inventory hash table read as side file) HDFS Hive : Dynamically Partitioned Hash Join

40 © Hortonworks Inc Tez – Bringing it all together Page 40 Architecting the Future of Big Data Tez Session populates container pool Dimension table calculation and HDFS split generation in parallel Dimension tables broadcasted to Hive MapJoin tasks Final Reducer pre- launched and fetches completed inputs TPCDS – Query-27 with Hive on Tez

41 © Hortonworks Inc Tez – Bridging the Data Spectrum Page 41 Fact Table Dimension Table 1 Result Table 1 Dimension Table 2 Result Table 2 Dimension Table 3 Result Table 3 Broadcast Join Shuffle Join Typical pattern in a TPC-DS query Fact Table Dimension Table 1 Broadcast join for small data sets Based on data size, the query optimizer can run either plan as a single Tez job Broadcast Join

42 © Hortonworks Inc Tez – Current status Apache Project –Growing community of contributors and users –Latest release is Focus on stability –Testing and quality are highest priority –Code ready and deployed on multi-node environments at scale Support for a vast topology of DAGs – Already functionally equivalent to Map Reduce. Existing Map Reduce jobs can be executed on Tez with few or no changes –Rapid adoption by important open source projects and by vendors Page 42

43 © Hortonworks Inc Tez – Developer Release API stability – Intuitive and clean APIs that will be supported and be backwards compatible with future releases Local Mode Execution – Allows the application to run entirely in the same JVM without any need for a cluster. Enables easy debugging using IDE’s on the developer machine Performance Debugging – Adds swim-lanes analysis tool to visualize job execution and help identify performance bottlenecks Documentation and tutorials to learn the APIs Page 43

44 © Hortonworks Inc Tez – Adoption Apache Hive Hadoop standard for declarative access via SQL-like interface (Hive 13) Apache Pig Hadoop standard for procedural scripting and pipeline processing (Pig 14) Cascading Developer friendly Java API and SDK Scalding (Scala API on Cascading) (3.0) Commercial Vendors ETL : Use Tez instead of MR or custom pipelines Analytics Vendors : Use Tez as a target platform for scaling parallel analytical tools to large data-sets Page 44

45 © Hortonworks Inc Tez – Adoption Path Pre-requisite : Hadoop 2 with YARN Tez has zero deployment pain. No side effects or traces left behind on your cluster. Low risk and low effort to try out. Using Hive, Pig, Cascading, Scalding Try them with Tez as execution engine Already have MapReduce based pipeline Use configuration to change MapReduce to run on Tez by setting ‘’ to ‘yarn-tez’ in mapred-site.xml Consolidate MR workflow into MR-DAG on Tez Change MR-DAG to use more efficient Tez constructs Have custom pipeline Wrap custom code/scripts into Tez inputs-processor-outputs Translate your custom pipeline topology into Tez DAG Change custom topology to use more efficient Tez constructs Page 45

46 © Hortonworks Inc Tez – Roadmap Richer DAG support – Addition of vertices at runtime – Shared edges for shared outputs – Enhance Input/Output library Performance optimizations – Improve support for high concurrency – Improve locality aware scheduling – Add framework level data statistics – HDFS memory storage integration Usability – Stability and testability – UI integration with YARN – Tools for performance analysis and debugging Page 46

47 © Hortonworks Inc Tez – Community Early adopters and code contributors welcome –Adopters to drive more scenarios. Contributors to make them happen. Tez meetup for developers and users – Technical blog series – processing processing Useful links –Work tracking: –Code: – Developer list: User list: Issues list: Page 47

48 © Hortonworks Inc Tez – Takeaways Distributed execution framework that models processing as dataflow graphs Customizable execution architecture designed for extensibility and user defined performance optimizations Works out of the box with the platform figuring out the hard stuff Zero install – Safe and Easy to Evaluate and Experiment Addresses common needs of a broad spectrum of applications and usage patterns Open source Apache project – your use-cases and code are welcome It works and is already being used by Apache Hive and Pig Page 48

49 © Hortonworks Inc Tez Thanks for your time and attention! Video with Deep Dive on Tez 82d626867fe57de70b218c74c03e4 82d626867fe57de70b218c74c03e4 – WordCount code walk through with 0.5 APIs. Debugging in local mode. Execution and profiling in cluster mode. (Starts at ~27 min. Ends at ~37 min) Page 49

Download ppt "© Hortonworks Inc. 2013 Page 1 Apache Tez : Accelerating Hadoop Data Processing Bikas"

Similar presentations

Ads by Google