.png){kind=icon}
Return
HPCBL: A Privacy-Preserving Data Computing Model for the Supercomputing Internet
.png){kind=icon}
PRE
AI
DOI：10.1109/TCC.2025.3604606.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
HPC-cloud is becoming popular as it allows supercomputers to provide computing service with parallelism and high performance based on public cloud technology. Moreover, supercomputers across organizations are forming a network to scale the computational and storage capability of their service. However, the distributed computing process in the supercomputer internet requires a large amount of data transmission and access, arousing privacy and control problems. In this paper, we propose HPCBL, a blockchain-enabled decentralized data computing architecture that ensures private data sharing and computing in the Supercomputer Internet. We also propose a decentralized authentication scheme that entitles users the full control of their anonymous identity to support user private interactions with the Supercomputing Internet. The scheme supports anonymous self-derivative credentials for pair-wised access control and user-optional accountability without a trusted arbiter. We implemented a HPCBL prototype to empirically assess its performance.
Keywords:
Blockchains
Authentication
Security
Supercomputers
Computer architecture
Cloud computing
Computational modeling
Privacy
Data models
Federated learning
HPC-cloud
anonymous authentication
blockchain
privacy-preserving data computing
user-centric
redactable signatures
Journal
I
IF:
5
Papers:
1.9K
Citations:
4.3K
Organization
Cited Papers
.png){kind=icon}
ShareSaveShareSaveAn Efficient System for Non-transferable Anonymous Credentials with Optional Anonymity Revocation
OAAI
OAAIShareSaveShareSave
View more