A Fast Updatable Implementation of Index Generation Functions Using Multiple IGUs

• SASAO Tsutomu

  Department of Computer Science, Meiji University

Abstract

This paper presents a method to realize index generation functions using multiple Index Generation Units (IGUs). The architecture implements index generation functions more efficiently than a single IGU when the number of registered vectors is very large. This paper proves that independent linear transformations are necessary in IGUs for efficient realization. Experimental results confirm this statement. Finally, it shows a fast update method to IGUs.

Journal

• IEICE Transactions on Information and Systems

  IEICE Transactions on Information and Systems E100.D (8), 1574-1582, 2017

  The Institute of Electronics, Information and Communication Engineers

Keywords

• random function
• CAM
• content-addressable memory
• linear decomposition
• linear transformation
• statistical analysis
• update method

Details

• CRID: 1390282679354093184
• NII Article ID: 130005876098
• DOI: 10.1587/transinf.2016lop0001
• ISSN: 17451361
• ISSN: 09168532
• Web Site: https://www.jstage.jst.go.jp/article/transinf/E100.D/8/E100.D_2016LOP0001/_pdf
• Text Lang: en
• Data Source:
  • JaLC
  • Crossref
  • CiNii Articles
  • KAKEN
• Abstract License Flag: Disallowed