Modern Information Retrieval

1
Modern Information Retrieval

• Chapter 9 Parallel and Distributed IR
• Section 9.1 Introduction
• Section 9.2.2. MIMD Architectures
• Inverted Files
• November 5, 1999

2
Summary

• Introduction
• Review of parallel computing and parallel program
  performance measures
• Exploration of techniques for implementing
  inverted file on MIMD parallel architecture
• Conclusion

3
Introduction

• The volume of electronic text available online
  today is staggering.
• The WWW contains over 800 millions pages of text,
  comprising nearly 6 terabytes of data (NATUREVol
  4008 July 1999www.nature.com).
• As document collections grow larger, they become
  more expensive to manage with an information
  retrieval system.
• To support the demanding requirements of modern
  search environments, we must turn to alternative
  architectures and algorithms.

4
Parallel Computing

• Parallel computing is the simultaneous aplication
  of multiple processors to solve a single problem.
• Flynns Taxonomy
• SISD single instruction, single data
• SIMD single instruction, multiple data
• MISD multiple instruction, single data
• MIMD multiple instruction, multiple data

5
Parallel Program Performance Measures

• Speedup
• Amdahls Law
• where f is the fraction of the problem that
• must be computed sequencially
• N is the number of processors.

6
Parallel Program Performance Measures

• Efficiency
• where S is speedup
• N is the number of processors.

7
MIMD Architectures

• MIMD architectures offer a great deal of
  flexibility in how parallelism is defined and
  exploited to solve a problem.
• There are two ways in which a retrieval system
  can exploit a MIMD machine
• Parallel multitasking
• Partitioned parallel processing.

8
MIMD Architectures

• Parallel multitasking on a MIMD machine

9
MIMD Architectures
Partitioned parallel processing on a MIMD machine
10
MIMD Architectures
Basic data elements processed by a seach algorithm
11
MIMD Architectures

• There are two possible methods for partitioning
  the data
• Document partitioning the N documents are
  distributed across the P processors each
  parallel process evaluates the query on the
  subcollection of N/P documents assigned to it
• Term partitioning the t indexing items are
  distributed across the P processors the
  evaluation process for each document is spread
  over multiple processors.

12
Inverted FilesLogical Document Partitioning

• Data Partitioning
• The data partitioning is done logically using
  essentially the same basic underlying inverted
  file index as in the original sequential
  algorithm
• The inverted file is extended to give each
  parallel process direct access to that portion of
  the index related to the processors
  subcollection of documents.

13
Inverted FilesLogical Document Partitioning
Extended dictionary entry for document partitionin
g
14
Inverted FilesLogical Document Partitioning

• Query Evaluation
• The broker initiates P parallel processes to
  evaluate the query
• Each process executes the same document scoring
  algorithm on its document subcollection
• The search processes record document scores in a
  single shared array of document score
  accumulators
• The broker produces the final ranked list of
  documents.

15
Inverted Files Logical Document Partitioning

• Inverted File Construction
• The indexer partitions the documents among the
  processors
• Each indexing process generates a batch of
  inverted lists, sorted by indexing item
• A merge step is performed to create the final
  inverted file.

16
Inverted FilesPhysical Document Partitioning

• Data Partitioning
• The documents are physically partitioned into
  separate subcollections, one for each parallel
  processor
• Each subcollection has its own inverted file.

17
Inverted FilesPhysical Document Partitioning

• Query Evaluation
• The broker distributes the query to all of the
  parallel search processes
• Each parallel search process evaluates the query
  on its portion of the document collection,
  producing an intermediate hit-list
• The broker collects the intermediate hit-lists
  from all of the parallel search processes and
  merges them into a final hit-list.

18
Inverted FilesPhysical Document Partitioning

• Inverted File Construction
• Each processor creates, in parallel, its own
  complete index corresponding to its document
  partition
• A merge step is performed to accumulate the
  global statistics for all of the partitions and
  distribute them to each of the partition
  dictionaries.

19
Inverted FilesTerm Partitioning

• Data Partitioning
• Inverted lists are spread across the processors.

20
Inverted FilesTerm Partitioning

• Query Evaluation
• Query is decomposed into indexing items and each
  indexing item is sent to the processor that holds
  the corresponding inverted list
• The processors create hit-lists with partial
  document scores and return them to the broker
• The broker combines the hit-lists.

21
Inverted FilesTerm Partitioning

• Inverted File Construction
• Inverted file is created using the parallel
  construction technique described for logical
  document partitioning.

22
Example
Document collection Document
Text 1
Pease porridge hot 2 Pease
porridge cold 3 Pease porridge
in the pot 4 Pease porridge
hot, pease porridge not cold 5
Pease porridge cold, pease porridge not hot
6 Pease porridge hot in the pot
23
Example
Inverted File
24
Example
Logical Document Partitioning
25
Example
Physical Document Partitioning
26
Example
Term Partitioning
27
Conclusion

• The task of indexing and searching in very large
  text collections is costly
• Faster indexing and searching algorithms are
  always desirable and the use of parallel hardware
  is and obvious alternative
• We discussed two possible organization for the
  document collection index on a MIMD parallel
  architecture
• Document partitioning
• Term partitioning.

28
Conclusion

• Document partitioning affords simpler inverted
  index construction and maintenance than term
  partitioning
• When term distributions in the documents and
  queries are more skewed, document partitioning
  performs better
• When terms are uniformily distributed in user
  queries, term partitioning performs better.

29
Adicional References
Lawrence, S., Giles, C.L. 1999. Accessibility of
Information on the Web. Nature.
Vol.400.pp.107-109. Ribeiro-Neto, B.A., Barbosa,
R.A. 1998. Query Performance for Tighly Coupled
Distributed Digital Libraries. Digital Libraries
98. pp.182-190. Ribeiro-Neto, B.A., Moura, E.S.,
Neubert, M.S., Ziviani, N. 1999. Efficient
Distributed Algorithms to Build Inverted Files.
SIGIR99. pp.105-112.