Empirical Study of Blockchains
Smart contracts
Massimo Bartoletti and Livio Pompianu (University of Cagliari)
WTSC'17
https://gyazo.com/6394f81416b21b6566a7798d963a5b29
University of Cagliari
https://gyazo.com/a718951f0030f8fcd9e9139761dabdcchttps://gyazo.com/be3d04fe0b0b71e3436593853d65ea10
Andra Anoaica, Hugo Levard (IRT SystemX)
NTMS'18
https://gyazo.com/d52c7d767cf01eb4e9124510c7329de8https://gyazo.com/dae200dae4f9368c2cfe98bc9169abc5
Stefano Ferretti, Gabriele D’Angelo (University of Bologna)
In Concurrency and Computation Practice and Experience
https://gyazo.com/17490ab1365f0e7c8d6ac4ed3b5e72b4
Lucianna Kiffer (Northeastern University), Dave Levin (University of Maryland), Alan Mislove (Northeastern University)
IMC'18
Findings (using customized geth client):
Contracts are three times more likely to be created by other contracts than they are by users
Over 60% of contracts have never been interacted with
Less than 10% of user created contracts are unique, and less than 1% of contract-created contracts are so (implies substantial code re-use in Ethereum)
Monika di Angelo, Gernot Salzer (TU Wien)
BCC'19
Data: by the end of 2018, from customized Parity client
63% of the contracts are never called
Non-deployment (of code): The aim of creation is rather the execution of the deployment code itself
Loners (never called contracts but w/ code) comprise 46%. However, many of them will probably be called in the future (like GasToken and wallets).
Mayflies (contracts that selfdestruct in the same transaction as they are created): 1.9M
Predominant pattern: Token Harvesting, hoarding of airdropped tokens by dozens of mayflies, , escaping blacklisting
Pass the "is human" test since they have no code
Breeders (contracts creating >1K contracts): only 276 breeders created 8.7 M contracts (out of 11.1M contracts created)
Responsible for 94% of the mayflies, 2.4M wallets, 1M gastokens, 0.4M ENS.
Code reuse: 11.1M contracts → only 94k (~0.1M) different code skeletons
Busy Bees (contracts sending and receiving over 1k calls) are rare but account for over 60% of all messages. In contrast, the overwhelming majority of contracts (more than 99%) only has 10 interactions or less.
OTM'19
Analyze the usage of SafeMath library
Tokens
Friedhelm Victor and Bianca Katharina L¨uders (TU Berlin)
FC'19
MIT
Users' Address
Friedhelm Victor (Technical University of Berlin)
FC'20
Contribution: The first clustering heuristics for Ethereum’s account
1. Exchanges' deposit address: Exhanges typically creates a deposit address per custormer to credit the assets
Multiple addresses that send funds to the same deposit address are highly likely to be controlled by the same entity
2. Airdrop multi-participation: Multiple addresses receives airdrops send them to the same address
3. ERC20 Self approval for test purposes or risk distribution over several addresses
Result: cluster 17.9% of all active externally owned account addresses
Drawback (of clustering in general): difficult to assess the quality of the clustering heuristics
Price
Yitao Li (Purdue University) et al.
SDM'20
Relationships between transaction graph properties and crypto price dynamics
Fudan University
The relationship between user-to-user interactions and price