This paper investigates the performance interference of blockchain services that run on cloud data centers. As the data centers offer shared computing resources to multiple services, the blockchain services can experience performance interference due to the co-located services. We explore the impact of the interference on Fabric performance and develop a new technique to offer performance isolation for Hyperledger Fabric, the most popular blockchain platform. First, we analyze the characteristics of the different components in Hyperledger Fabric and show that Fabric components have different impacts on the performance of Fabric. Then, we present QiOi, component-level performance isolation technique for Hyperledger Fabric. The key idea of QiOi is to dynamically control the CPU scheduling of Fabric components to cope with the performance interference. We im-plement QiOi as a user-level daemon and evaluate how QiOi mitigates the performance interference of Fabric. The evaluation results demonstrate that QiOi mitigates performance degradation of Fabric by 22% and improves Fabric latency by 2.5 times without sacrificing the performance of co-located services. In addition, we show that QiOi can support different ordering services and chaincodes with negligible overhead to Fabric performance.
Bibliographical noteFunding Information:
Funding: This work was partly supported by Institute of Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT) (No. 2015-0-00280 (SW Starlab) Next generation cloud infra-software toward the guarantee of performance and security SLA). This research was also supported by Next Generation Engineering Researcher Program of National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT (No. NRF-2019H1D8A2105513).
© 2021 by the authors. Licensee MDPI, Basel, Switzerland.
- Hyperledger Fabric
- Performance interference
- Performance isolation, cloud computing
- Private blockchain
ASJC Scopus subject areas
- General Materials Science
- General Engineering
- Process Chemistry and Technology
- Computer Science Applications
- Fluid Flow and Transfer Processes