Apache Hadoop
Hadoop Logo | |
Developer(s) | Apache Software Foundation |
---|---|
Stable release | 0.21
/ August 23, 2010 |
Repository | |
Written in | Java |
Operating system | Cross-platform |
Type | Distributed File System |
License | Apache License 2.0 |
Website | http://hadoop.apache.org/ |
Apache Hadoop is a software framework that supports data-intensive distributed applications under a free license.[1] It enables applications to work with thousands of nodes and petabytes of data. Hadoop was inspired by Google's MapReduce and Google File System (GFS) papers.
Hadoop is a top-level Apache project being built and used by a global community of contributors,[2] using the Java programming language. Yahoo! has been the largest contributor[3] to the project, and uses Hadoop extensively across its businesses.[4]
Hadoop was created by Doug Cutting,[5] who named it after his son's stuffed elephant.[6] It was originally developed to support distribution for the Nutch search engine project.[7]
Architecture
Hadoop consists of the Hadoop Common, which provides access to the filesystems supported by Hadoop. The Hadoop Common package contains the necessary jar files and scripts needed to start Hadoop. The package also provides source code, documentation, and a contribution section which includes projects from the Hadoop Community.
A key feature is that for effective scheduling of work, every filesystem should provide location awareness: the name of the rack (more precisely, of the network switch) where a worker node is. Hadoop applications can use this information to run work on the node where the data is, and, failing that, on the same rack/switch, so reducing backbone traffic. The HDFS filesystem uses this when replicating data, to try to keep different copies of the data on different racks. The goal is to reduce the impact of a rack power outage or switch failure so that even if these events occur, the data may still be readable.[8]
![Hadoop cluster](http://upload.wikimedia.org/wikipedia/commons/thumb/2/2b/Hadoop_1.png/250px-Hadoop_1.png)
A typical Hadoop cluster will include a single master and multiple slave nodes. The master node consists of a jobtracker, tasktracker, namenode, and datanode. A slave or compute node consists of a datanode and tasktracker. Hadoop requires JRE 1.6 or higher and ssh to be set up between nodes in the cluster.
Filesystems
Hadoop Distributed File System
The HDFS is a distributed, scalable, and portable filesystem written in Java for Hadoop. Each node in an Hadoop instance typically has a single datanode; a cluster of datanodes form the HDFS cluster. The situation is typical because each node does not require a datanode to be present. Each datanode serves up blocks of data over the network using a block protocol specific to HDFS. The filesystem uses the TCP/IP layer for communication; clients use RPC to communicate between each other. The HDFS stores large files (an ideal file size is a multiple of 64 MB[9]), across multiple machines. It achieves reliability by replicating the data across multiple hosts, and hence does not require RAID storage on hosts. With the default replication value, 3, data is stored on three nodes: two on the same rack, and one on a different rack. Data nodes can talk to each other to rebalance data, to move copies around, and to keep the replication of data high. The HDFS is not fully POSIX compliant because the requirements for a POSIX filesystem differ from the target goals for a Hadoop application. The tradeoff of not having a fully POSIX compliant filesystem is increased performance for data throughput. The HDFS was designed to handle very large files.[9] The HDFS does not provide High Availability.
A filesystem requires one unique server, the name node. This is a single point of failure for an HDFS installation. If the name node goes down, the filesystem is offline. When it comes back up, the name node must replay all outstanding operations. This replay process can take over half an hour for a big cluster.[10] The filesystem includes what is called a Secondary Namenode, which misleads some people into thinking that when the Primary Namenode goes offline, the Secondary Namenode takes over. In fact, the Secondary Namenode regularly connects with the Primary Namenode and builds snapshots of the Primary Namenode's directory information, which is then saved to local/remote directories. These checkpointed images can be used to restart a failed Primary Namenode without having to replay the entire journal of filesystem actions, the edit log to create an up-to-date directory structure.
An advantage of using the HDFS is data awareness between the jobtracker and tasktracker. The jobtracker schedules map/reduce jobs to tasktrackers with an awareness of the data location. An example of this would be if node A contained data (x,y,z) and node B contained data (a,b,c). The jobtracker will schedule node B to perform map/reduce tasks on (a,b,c) and node A would be scheduled to perform map/reduce tasks on (x,y,z). This reduces the amount of traffic that goes over the network and prevents unnecessary data transfer. When Hadoop is used with other filesystems this advantage is not available. This can have a significant impact on the performance of job completion times, which has been demonstrated when running data intensive jobs.[11]
Another limitation of HDFS is that it cannot be directly mounted by an existing operating system. Getting data into and out of the HDFS file system, an action that often needs to be performed before and after executing a job, can be inconvenient. A Filesystem in Userspace (FUSE) virtual file system has been developed to address this problem, at least for Linux and some other Unix systems.
File access can be achieved through the native Java API, the Thrift API to generate a client in the language of your choosing (C++, Java, Python, PHP, Ruby, Erlang, Perl, Haskell, C#, Cocoa, Smalltalk, and OCaml), the command line interface, or browsed through the HDFS-UI webapp over HTTP.
Other filesystems
By June 2010, the list of supported filesystems included:
- HDFS: Hadoop's own rack-aware filesystem. This is designed to scale to tens of petabytes of storage and runs on top of the filesystems of the underlying operating systems.
- Amazon S3 filesystem. This is targeted at clusters hosted on the Amazon Elastic Compute Cloud server-on-demand infrastructure. There is no rack-awareness in this filesystem, as it is all remote.
- CloudStore (previously Kosmos Distributed File System), which is rack-aware.
- FTP Filesystem: this stores all its data on remotely accessible FTP servers.
- Read-only HTTP and HTTPS file systems.
Hadoop can work directly with any distributed file system which can be mounted by the underlying operating system simply by using a file:// URL, however this comes at a price: the loss of locality. To reduce network traffic, Hadoop needs to know which servers are closest to the data; this is information which Hadoop-specific filesystem bridges can provide.
Out-of-the-box, this includes Amazon S3, and the CloudStore filestore, through s3:// and kfs:// URLs directly.
A number of third party filesystem bridges have also been written, none of which are currently in Hadoop distributions. These may offer superior availability or scalability, and possibly a more general-purpose filesystem than HDFS, which is biased towards large files and only offers a subset of the expected semantics of a Posix Filesystem: no locking or writing to anywhere other than the tail of a file.
- In 2009 IBM discussed running Hadoop over the IBM General Parallel File System.[12] The source code was published in October 2009.[13]
- In April 2010, Parascale published the source code to run Hadoop against the Parascale filesystem.[14]
- In April 2010, Appistry, Inc. announced and released a Hadoop filesystem driver for use with its own CloudIQ Storage product.[15]
- In June 2010, HP discussed a location-aware IBRIX Fusion filesystem driver.[16]
Job Tracker and Task Tracker: the MapReduce engine
Above the file systems comes the MapReduce engine, which consists of one Job Tracker, to which client applications submit MapReduce jobs. The Job Tracker pushes work out to available Task Tracker nodes in the cluster, striving to keep the work as close to the data as possible. With a rack-aware filesystem, the Job Tracker knows which node contains the data, and which other machines are nearby. If the work cannot be hosted on the actual node where the data resides, priority is given to nodes in the same rack. This reduces network traffic on the main backbone network. If a Task Tracker fails or times out, that part of the job is rescheduled. The TaskTracker on each node spawns off a separate JVM process to prevent the TaskTracker itself from failing if the running job has problems. A heartbeat is sent from the TaskTracker to the JobTracker every few minutes to its status. The Job Tracker and TaskTracker status an information can be viewed from a web browser which is powered by Jetty.
If the Job Tracker failed on Hadoop 0.20 or earlier, all ongoing work was lost. Hadoop version 0.21 added some checkpointing to this process; the Job Tracker records what it is up to in the filesystem. When a Job Tracker starts up, it looks for any such data, so that it can restart work from where it left off. In earlier versions of Hadoop, all active work was lost when a Job Tracker restarted.
Known limitations of this approach are:
- The allocation of work to task trackers is very simple. Every task tracker has a number of available slots (such as "4 slots"). Every active map or reduce task takes up one slot. The Job Tracker allocates work to the tracker nearest to the data with an available slot. There is no consideration of the current active load of the allocated machine, and hence its actual availability.
- If one task tracker is very slow, it can delay the entire MapReduce operation -especially towards the end of a job, where everything can end up waiting for a single slow task. With speculative-execution enabled, however, a single task can be executed on multiple slave nodes. So this technology makes the networking very efficient.
Scheduling
By default Hadoop uses FIFO to schedule jobs from a work queue. In version 0.19 the job scheduler was refactored out of the Hadoop which added the ability to use an alternate scheduler.[17]
Fair scheduler
The fair scheduler was developed by Facebook. The goal of the fair scheduler is to provide fast response times for small jobs and QOS for production jobs. The fair scheduler has three basic concepts.[18]
- Jobs are grouped into Pools.
- Each pool is assigned a guaranteed minimum share.
- Excess capacity is spilt between jobs.
By default jobs that are uncategorized go into a default pool. Pools have to specify the minimum number of map slots, reduce slots, and a limit on the number of running jobs.
Capacity scheduler
The capacity scheduler was developed by Yahoo. The capacity scheduler supports several features which are similar to the fair scheduler.[19]
- Jobs are submitted into queues.
- Queues are allocated a fraction of the total resource capacity.
- Free resources are allocated to queues beyond their total capacity.
- Within a queue a job with a high level of priority will have access to the queue's resources.
There is no preemption once a job is running.
Other applications
The HDFS filesystem is not restricted to MapReduce jobs. It can be used for other applications, many of which are under development at Apache. The list includes the HBase database, the Apache Mahout machine learning system, and matrix operations. Hadoop can in theory be used for any sort of work that is batch-oriented rather than real-time, that is very data-intensive, and able to work on pieces of the data in parallel. As of October 2009, commercial applications of Hadoop[20] included:
- Log and/or clickstream analysis of various kinds
- Marketing analytics
- Machine learning and/or sophisticated data mining
- Image processing
- Processing of XML messages
- Web crawling and/or text processing
- General archiving, including of relational/tabular data, e.g. for compliance
Prominent users
Hadoop at Yahoo!
On February 19, 2008, Yahoo! launched what it claimed was the world's largest Hadoop production application. The Yahoo! Search Webmap is a Hadoop application that runs on more than 10,000 core Linux cluster and produces data that is now used in every Yahoo! Web search query.[21]
There are multiple Hadoop clusters at Yahoo!, each occupying a single datacenter (or fraction thereof). No HDFS filesystems or MapReduce jobs are split across multiple datacenters; instead each datacenter has a separate filesystem and workload. The cluster servers run Linux, and are configured on boot using Kickstart. Every machine bootstraps the Linux image, including the Hadoop distribution. Work that the clusters perform is known to include the index calculations for the Yahoo! search engine.
On June 10, 2009, Yahoo! made available the source code to the version of Hadoop it runs in production.[22] While Yahoo! contributes back all work it does on Hadoop to the open-source community, the company's developers also fix bugs and provide stability improvements internally, and release this patched source code so that other users may benefit from their effort.
Other users
Besides Yahoo!, many other organizations are using Hadoop to run large distributed computations. Some of the notable users include:[2]
Hadoop on Amazon EC2/S3 services
It is possible to run Hadoop on Amazon Elastic Compute Cloud (EC2) and Amazon Simple Storage Service (S3).[24] As an example The New York Times used 100 Amazon EC2 instances and a Hadoop application to process 4TB of raw image TIFF data (stored in S3) into 11 million finished PDFs in the space of 24 hours at a computation cost of about $240 (not including bandwidth).[25]
There is support for the S3 filesystem in Hadoop distributions, and the Hadoop team generates EC2 machine images after every release. From a pure performance perspective, Hadoop on S3/EC2 is inefficient, as the S3 filesystem is remote and delays returning from every write operation until the data are guaranteed to not be lost. This removes the locality advantages of Hadoop, which schedules work near data to save on network load.
On April 2, 2009 Amazon announced the beta release of a new service called Amazon Elastic MapReduce which they describe as "a web service that enables businesses, researchers, data analysts, and developers to easily and cost-effectively process vast amounts of data. It utilizes a hosted Hadoop framework running on the web-scale infrastructure of Amazon Elastic Compute Cloud (Amazon EC2) and Amazon Simple Storage Service (Amazon S3)."[26]
Hadoop at Google and IBM
IBM and Google announced an initiative in 2007 to use Hadoop to support university courses in distributed computer programming.[27]
In 2008 this collaboration, the Academic Cloud Computing Initiative (ACCI), partnered with the National Science Foundation to provide grant funding to academic researchers interested in exploring large-data applications. This resulted in the creation of the Cluster Exploratory (CLuE) program.[28]
Hadoop with Sun Grid Engine
Hadoop can also be used in compute farms and high-performance computing environments. Integration with Sun Grid Engine was released, and running Hadoop on Sun Grid (Sun's on-demand utility computing service) is possible.[29] In the initial implementation of the integration, the CPU-time scheduler has no knowledge of the locality of the data. A key feature of the Hadoop Runtime, "do the work in the same server or rack as the data" is therefore lost.
During the Sun HPC Software Workshop '09, an improved integration with data-locality awareness was announced.[30]
Sun also has the Hadoop Live CD OpenSolaris project, which allows running a fully functional Hadoop cluster using a live CD.[31]
See also
- Nutch - an effort to build an open source search engine based on Lucene and Hadoop. Also created by Doug Cutting.
- Datameer Analytics Solution (DAS) – data source integration, storage, analytics engine and visualization
- HBase - BigTable-model database.
- Hypertable - HBase alternative
- MapReduce - Hadoop's fundamental data filtering algorithm
- Apache Mahout - Machine Learning algorithms implemented on Hadoop
- Apache Cassandra - A column-oriented database that supports access from Hadoop
- Cloud computing
- Big data
References
- ^ "Hadoop is a framework for running applications on large clusters of commodity hardware. The Hadoop framework transparently provides applications both reliability and data motion. Hadoop implements a computational paradigm named map/reduce, where the application is divided into many small fragments of work, each of which may be executed or re-executed on any node in the cluster. In addition, it provides a distributed file system that stores data on the compute nodes, providing very high aggregate bandwidth across the cluster. Both map/reduce and the distributed file system are designed so that node failures are automatically handled by the framework." Hadoop Overview
- ^ a b Applications and organizations using Hadoop
- ^ Hadoop Credits Page
- ^ Yahoo! Launches World's Largest Hadoop Production Application
- ^ Hadoop creator goes to Cloudera
- ^ Ashlee Vance (2009-03-17). "Hadoop, a Free Software Program, Finds Uses Beyond Search". New York Times. Retrieved 2010-01-20.
- ^ "Hadoop contains the distributed computing platform that was formerly a part of Nutch. This includes the Hadoop Distributed Filesystem (HDFS) and an implementation of map/reduce." About Hadoop
- ^ http://hadoop.apache.org/core/docs/r0.17.2/hdfs_user_guide.html#Rack+Awareness
- ^ a b The Hadoop Distributed File System: Architecture and Design
- ^ Improve Namenode startup performance. "Default scenario for 20 million files with the max Java heap size set to 14GB : 40 minutes. Tuning various Java options such as young size, parallel garbage collection, initial Java heap size : 14 minutes"
- ^ [1] Improving MapReduce Performance through Data Placement in Heterogeneous Hadoop Clusters April 2010
- ^ ", "Cloud analytics: Do we really need to reinvent the storage stack?"" (PDF). IBM. 2009-06.
{{cite web}}
: Check date values in:|date=
(help) - ^ "HADOOP-6330: Integrating IBM General Parallel File System implementation of Hadoop Filesystem interface". IBM. 2009-10-23.
- ^ "HADOOP-6704: add support for Parascale filesystem". Parascale. 2010-04-14.
- ^ "Replace HDFS with CloudIQ Storage". Appistry,Inc. 2010-07-06.
- ^ "High Availability Hadoop". HP. 2010-06-09.
- ^ [ https://issues.apache.org/jira/browse/HADOOP-3412] #HADOOP-3412 Refactor the scheduler out of the JobTracker - ASF JIRA
- ^ [2] Hadoop Fair Scheduler Design Document
- ^ [3] Capacity Scheduler Guide
- ^ "How 30+ enterprises are using Hadoop", in DBMS2
- ^ Yahoo! Launches World's Largest Hadoop Production Application (Hadoop and Distributed Computing at Yahoo!)
- ^ Hadoop and Distributed Computing at Yahoo!
- ^ "HBase at StumbleUpon". Retrieved 2010-06-26.
- ^ http://aws.typepad.com/aws/2008/02/taking-massive.html Running Hadoop on Amazon EC2/S3
- ^ Gottfrid, Derek (November 1, 2007). "Self-service, Prorated Super Computing Fun!". The New York Times. Retrieved May 4, 2010.
- ^ Amazon Elastic MapReduce Beta
- ^ Google Press Center: Google and IBM Announce University Initiative to Address Internet-Scale Computing Challenges
- ^ NSF, Google, IBM form CLuE
- ^ "Creating Hadoop pe under SGE". Sun Microsystems. 2008-01-16.
- ^ "HDFS-Aware Scheduling With Grid Engine". Sun Microsystems. 2009-09-10.
- ^ "OpenSolaris Project: Hadoop Live CD". Sun Microsystems. 2008-08-29.
Bibliography
- Lam, Chuck (July 28, 2010). Hadoop in Action (1st ed.). Manning Publications. p. 325. ISBN 1935182196.
- Venner, Jason (June 22, 2009). Pro Hadoop (1st ed.). Apress. p. 440. ISBN 1430219424.
- White, Tom (June 16, 2009). Hadoop: The Definitive Guide (1st ed.). O'Reilly Media. p. 524. ISBN 0596521979.