MapReduce and the New Software Stack CHAPTER 2 1.
-
Upload
sharon-welch -
Category
Documents
-
view
223 -
download
0
Transcript of MapReduce and the New Software Stack CHAPTER 2 1.
![Page 1: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/1.jpg)
1
MapReduce and the New Software StackCHAPTER 2
![Page 2: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/2.jpg)
2
Single Node Architecture
J. LESKOVEC, A. RAJARAMAN, J. ULLMAN: MINING OF MASSIVE DATASETS, HTTP://WWW.MMDS.ORG
Memory
Disk
CPU
Machine Learning, Statistics
“Classical” Data Mining
![Page 3: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/3.jpg)
3
Motivation: Google Example
20+ billion web pages x 20KB = 400+ TB 1 computer reads 30-35 MB/sec from disk
~4 months to read the web ~1,000 hard drives to store the web Takes even more to do something useful
with the data!
J. LESKOVEC, A. RAJARAMAN, J. ULLMAN: MINING OF MASSIVE DATASETS, HTTP://WWW.MMDS.ORG
![Page 4: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/4.jpg)
4
Cluster Architecture
J. LESKOVEC, A. RAJARAMAN, J. ULLMAN: MINING OF MASSIVE DATASETS, HTTP://WWW.MMDS.ORG
Mem
Disk
CPU
Mem
Disk
CPU
…
Switch
Each rack contains 16-64 nodes
Mem
Disk
CPU
Mem
Disk
CPU
…
Switch
Switch1 Gbps between any pair of nodesin a rack
2-10 Gbps backbone between racks
![Page 5: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/5.jpg)
5
![Page 6: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/6.jpg)
6
Large-scale Computing A system has the more frequently something in the system will not be working at any given time.
the principal failure modes are the loss of a single node and the loss of an entire rack.
If we had to abort and restart the computation every time one component failed, then the computation might never complete successfully.
![Page 7: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/7.jpg)
7
Idea and SolutionIdea:
Files must be stored redundantly.Computations must be divided into tasks.
Solution:To exploit cluster computing, files must look and behave
somewhat differently from the conventional file systems found on single computers.
This new file system, often called a distributed file system or DFS.
![Page 8: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/8.jpg)
8
Typical usage pattern◦ Provides global file namespace◦ Huge files (100s of GB to TB)◦ Data is rarely updated in place◦ Reads and appends are common
![Page 9: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/9.jpg)
9
Distributed File System Chunk servers
◦ File is split into contiguous chunks◦ Typically each chunk is 16-64MB◦ Each chunk replicated (usually 2x or 3x)◦ Try to keep replicas in different racks
Master node◦ a.k.a. Name Node in Hadoop’s HDFS◦ Stores metadata about where files are stored◦ Might be replicated
Client library for file access◦ Talks to master to find chunk servers ◦ Connects directly to chunk servers to access data
J. LESKOVEC, A. RAJARAMAN, J. ULLMAN: MINING OF MASSIVE DATASETS, HTTP://WWW.MMDS.ORG
![Page 10: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/10.jpg)
10
Distributed File System
J. LESKOVEC, A. RAJARAMAN, J. ULLMAN: MINING OF MASSIVE DATASETS, HTTP://WWW.MMDS.ORG
Reliable distributed file system
Data kept in “chunks” spread across machines
Each chunk replicated on different machines
◦ Seamless recovery from disk or machine failure
C0 C1
C2C5
Chunk server 1
D1
C5
Chunk server 3
C1
C3C5
Chunk server 2
…C2D
0 D0
Bring computation directly to the data!
C0 C5
Chunk server N
C2D0
Chunk servers also serve as compute servers
![Page 11: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/11.jpg)
11
MapReduce MapReduce is a style of computing.
You can use an implementation of MapReduce to manage many large-scale computations in a way that is tolerant of hardware faults.
All you need to write are two functions, called Map and Reduce.
![Page 12: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/12.jpg)
12
MapReduce
![Page 13: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/13.jpg)
13
The Map Step Some number of Map tasks each are given one or more chunks from a distributed file system. These Map tasks turn the chunk into a sequence of key-value pairs.
The types of keys and values are each arbitrary.Further, keys are not “keys” in the usual sense; they do not have to be unique.
![Page 14: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/14.jpg)
14
MapReduce: The Map Step
J. LESKOVEC, A. RAJARAMAN, J. ULLMAN: MINING OF MASSIVE DATASETS, HTTP://WWW.MMDS.ORG
vk
k v
k v
mapvk
vk
…
k vmap
Inputkey-value pairs
Intermediatekey-value pairs
…
k v
map
![Page 15: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/15.jpg)
15
MapReduce: The Reduce Step The Reduce tasks work on one key at a time, and combine all the values associated with that key in some way.
These key-value pairs can be of a type different from those sent from Map tasks to Reduce tasks, but often they are the same type.
![Page 16: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/16.jpg)
16
MapReduce: The Reduce Step
J. LESKOVEC, A. RAJARAMAN, J. ULLMAN: MINING OF MASSIVE DATASETS, HTTP://WWW.MMDS.ORG
k v
…
k v
k v
k v
Intermediatekey-value pairs
Groupby key
reduce
reduce
k v
k v
k v
…
k v
…
k v
k v v
v v
Key-value groupsOutput key-value pairs
![Page 17: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/17.jpg)
17
MapReduce: The Reduce Step
Groupby key
reduce
![Page 18: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/18.jpg)
18
More Specifically Input: a set of key-value pairs Programmer specifies two methods:
◦ Map(k, v) <k’, v’>*◦ Takes a key-value pair and outputs a set of key-value pairs
◦ E.g., key is the filename, value is a single line in the file◦ There is one Map call for every (k,v) pair
◦ Reduce(k’, <v’>*) <k’, v’’>*◦ All values v’ with same key k’ are reduced together
and processed in v’ order◦ There is one Reduce function call per unique key k’
![Page 19: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/19.jpg)
19
Word CountingWarm-up task:
We have a huge text document
Count the number of times each distinct word appears in the file
Sample application: ◦ Analyze web server logs to find popular URLs
![Page 20: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/20.jpg)
20
MapReduce: Word Counting
The crew of the space shuttle Endeavor recently returned to Earth as ambassadors, harbingers of a new era of space exploration. Scientists at NASA are saying that the recent assembly of the Dextre bot is the first step in a long-term space-based man/mache partnership. '"The work we're doing now -- the robotics we're doing -- is what we're going to need ……………………..
Big document
(The, 1)(crew, 1)
(of, 1)(the, 1)
(space, 1)(shuttle, 1)(Endeavor,
1)(recently, 1)
….
(crew, 1)(crew, 1)(space, 1)
(the, 1)(the, 1)(the, 1)
(shuttle, 1)(recently, 1)
…
(crew, 2)(space, 1)
(the, 3)(shuttle, 1)(recently, 1)
…
MAP:Read input and produces a set
of key-value pairs
Group by key:
Collect all pairs with same key
Reduce:Collect all
values belonging to the key and
output
(key, value)
Provided by the programmer
(key, value)(key, value)
![Page 21: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/21.jpg)
J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive Datasets, http://www.mmds.org
21
Map-Reduce: EnvironmentMap-Reduce environment takes care of:
Partitioning the input data
Scheduling the program’s execution across a set of machines
Performing the group by key step
Handling machine failures
Managing required inter-machine communication
![Page 22: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/22.jpg)
J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive Datasets, http://www.mmds.org
22
Map-Reduce: A diagramBig document
MAP:Read input and
produces a set of key-value pairs
Group by key:Collect all pairs with same key
(Hash merge, Shuffle, Sort, Partition)
Reduce:Collect all values
belonging to the key and output
![Page 23: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/23.jpg)
J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive Datasets, http://www.mmds.org
23
Map-Reduce: In Parallel
All phases are distributed with many tasks doing the work
![Page 24: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/24.jpg)
J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive Datasets, http://www.mmds.org
24
Map-Reduce Programmer specifies:
◦ Map and Reduce and input files
Workflow:◦ Read inputs as a set of key-value-pairs◦ Map transforms input kv-pairs into a new set of
k'v'-pairs◦ Sorts & Shuffles the k'v'-pairs to output nodes◦ All k’v’-pairs with a given k’ are sent to the same
reduce◦ Reduce processes all k'v'-pairs grouped by key
into new k''v''-pairs◦ Write the resulting pairs to files
All phases are distributed with many tasks doing the work
Input 0
Map 0
Input 1
Map 1
Input 2
Map 2
Reduce 0 Reduce 1
Out 0 Out 1
Shuffle
![Page 25: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/25.jpg)
J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive Datasets, http://www.mmds.org
25
Data Flow Input and final output are stored on a distributed file system (FS):◦ Scheduler tries to schedule map tasks “close” to physical storage
location of input data
Intermediate results are stored on local FS of Map and Reduce workers
Output is often input to another MapReduce task
![Page 26: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/26.jpg)
J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive Datasets, http://www.mmds.org
26
Coordination: Master Master node takes care of coordination:
◦ Task status: (idle, in-progress, completed)◦ Idle tasks get scheduled as workers become available◦ When a map task completes, it sends the master the
location and sizes of its R intermediate files, one for each reducer
◦ Master pushes this info to reducers
Master pings workers periodically to detect failures
![Page 27: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/27.jpg)
J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive Datasets, http://www.mmds.org
27
Dealing with Failures Map worker failure
◦ Map tasks completed or in-progress at worker are reset to idle.
◦ Reduce workers are notified when task is rescheduled on another worker.
Reduce worker failure◦ Only in-progress tasks are reset to idle ◦ Reduce task is restarted
Master failure◦ MapReduce task is aborted and client is notified
![Page 28: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/28.jpg)
J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive Datasets, http://www.mmds.org
28
How many Map and Reduce jobs? M map tasks, R reduce tasks
Rule of a thumb:◦ Make M much larger than the number of nodes in the cluster◦ One DFS chunk per map is common◦ Improves dynamic load balancing and speeds up recovery
from worker failures
Usually R is smaller than M◦ Because output is spread across R files
![Page 29: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/29.jpg)
J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive Datasets, http://www.mmds.org
29
Task Granularity & Pipelining Fine granularity tasks: map tasks >> machines
◦ Minimizes time for fault recovery◦ Can do pipeline shuffling with map execution◦ Better dynamic load balancing
![Page 30: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/30.jpg)
J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive Datasets, http://www.mmds.org
30
Refinements: Backup Tasks Problem
◦ Slow workers significantly lengthen the job completion time:◦ Other jobs on the machine◦ Bad disks◦ Weird things
Solution◦ Near end of phase, spawn backup copies of tasks
◦ Whichever one finishes first “wins”
Effect◦ Dramatically shortens job completion time
![Page 31: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/31.jpg)
J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive Datasets, http://www.mmds.org
31
Refinement: Combiners Often a Map task will produce many pairs of the form (k,v1), (k,v2), … for the same key k
◦ E.g., popular words in the word count example
Can save network time by pre-aggregating values in the mapper:
◦ combine(k, list(v1)) v2
◦ Combiner is usually same as the reduce function
Works only if reduce function is commutative and associative
![Page 32: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/32.jpg)
J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive Datasets, http://www.mmds.org
32
Refinement: Combiners Back to our word counting example:
◦ Combiner combines the values of all keys of a single mapper (single machine):
◦ Much less data needs to be copied and shuffled!
![Page 33: MapReduce and the New Software Stack CHAPTER 2 1.](https://reader036.fdocuments.in/reader036/viewer/2022070415/5697bfa01a28abf838c950f8/html5/thumbnails/33.jpg)
J. Leskovec, A. Rajaraman, J. Ullman: Mining of Massive Datasets, http://www.mmds.org
33
Refinement: Partition Function
Want to control how keys get partitioned◦ Inputs to map tasks are created by contiguous splits of input file◦ Reduce needs to ensure that records with the same intermediate key end up at the
same worker
System uses a default partition function:◦ hash(key) mod R
Sometimes useful to override the hash function:◦ E.g., hash(hostname(URL)) mod R ensures URLs from a host end up
in the same output file