Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/105549
PIRA download icon_1.1View/Download Full Text
Title: Characterizing erasable accounts in ethereum
Authors: Li, X 
Chen, T
Luo, X 
Yu, J
Issue Date: 2020
Source: Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and lecture notes in bioinformatics), 2020, v. 12472, p. 352-371
Abstract: Being the most popular permissionless blockchain that supports smart contracts, Ethereum allows any user to create accounts on it. However, not all accounts matter. For example, the accounts due to attacks can be removed. In this paper, we conduct the first investigation on erasable accounts that can be removed to save system resources and even users’ money (i.e., ETH or gas). In particular, we propose and develop a novel tool named GLASER, which analyzes the State DataBase of Ethereum to discover five kinds of erasable accounts. The experimental results show that GLASER can accurately reveal 508,482 erasable accounts and these accounts lead to users wasting more than 106 million dollars. GLASER can help stop further economic loss caused by these detected accounts. Moreover, GLASER characterizes the attacks/behaviors related to detected erasable accounts through graph analysis.
Keywords: Ethereum
P2P system security
System maintainability
Publisher: Springer
Journal: Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and lecture notes in bioinformatics) 
ISBN: 978-3-030-62973-1
978-3-030-62974-8 (eBook)
ISSN: 0302-9743
EISSN: 1611-3349
DOI: 10.1007/978-3-030-62974-8_20
Description: 23rd International Conference, ISC 2020, Bali, Indonesia, December 16-18, 2020
Rights: © Springer Nature Switzerland AG 2020
This version of the proceeding paper has been accepted for publication, after peer review (when applicable) and is subject to Springer Nature’s AM terms of use(https://www.springernature.com/gp/open-research/policies/accepted-manuscript-terms), but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: http://dx.doi.org/10.1007/978-3-030-62974-8_20.
Appears in Collections:Conference Paper

Files in This Item:
File Description SizeFormat 
Li_Characterizing_Erasable_Accounts.pdfPre-Published version1.9 MBAdobe PDFView/Open
Open Access Information
Status open access
File Version Final Accepted Manuscript
Access
View full-text via PolyU eLinks SFX Query
Show full item record

Page views

117
Last Week
1
Last month
Citations as of Nov 30, 2025

Downloads

59
Citations as of Nov 30, 2025

SCOPUSTM   
Citations

6
Citations as of Dec 19, 2025

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.