The Bank of Korea: "CBDC Simulation System Operates Normally Even When Linked with Financial Institutions"

Processing Volume ↓ and Response Waiting Time ↑ Compared to Simulations

The Bank of Korea confirmed that the 'CBDC (Central Bank Digital Currency) simulation system' it established last year operates normally when linked with financial institutions. However, the number of processed transactions decreased, and response waiting times increased compared to the simulation conducted in a single cloud environment.

On the 8th, the Bank of Korea announced that through the linkage experiment, the already established CBDC simulation system functioned properly in an actual operational environment.

This linkage experiment was a research project conducted over about five months from July to December last year, with Crust as the main contractor, collaborating with a total of six companies (KPMG, Kakao Bank, Kakao Pay, Kakao Enterprise, and Engle). Fifteen financial institutions voluntarily expressed their intention to participate (including 14 banks such as Kookmin, Shinhan, Woori, Hana, Nonghyup, Busan, Daegu, Gyeongnam, Gwangju, Jeonbuk, Suhyup, Industrial Bank, Kakao, and K Bank, as well as the Korea Financial Telecommunications & Clearings Institute), and the selected institutions deployed IT systems and personnel during the experiment period.

The Bank of Korea confirmed that 64 key basic CBDC functions performed during the simulation operated normally even when linked with financial institutions.

In this experiment, the Bank of Korea and 12 participating institutions formed a distributed ledger network with a total of 15 nodes and conducted performance tests based on four scenarios (▲increase in transactions per second ▲expansion of simultaneous active users ▲reduction of transaction queue size ▲adjustment of block composition ratio). The experiment was conducted by registering 50 million users in the CBDC simulation system and measuring the processing performance of randomly input transactions over 30 minutes.

The results showed that the average transactions per second (TPS) was about 1,900, approximately 10% lower than last year's simulation result of around 2,100 TPS. However, the Bank of Korea explained that this 10% performance decrease is acceptable since it is higher than the average TPS peak of 1,200 in the domestic major small-value payment infrastructure, the Electronic Financial Network. Additionally, response waiting times increased by up to five times. As the IT system operating environments diversified, performance variations occurred among participating institutions.

Furthermore, the experiment confirmed that reducing the transaction queue size improved response waiting time delays, and increasing the block composition ratio enhanced throughput and average response time.

Moreover, if transaction processing errors occurred at a node of a specific institution, the remaining participating institutions continued their operations normally. Yoo Hee-jun, team leader of the Digital Currency Technology Team 1 at the Bank of Korea's Financial Clearing Department, stated, "Through this linkage experiment, we judged that the distributed ledger system has superior operational resilience compared to the existing centralized IT system, but we also found that management difficulties remain in communication and problem-solving methods among system managers." He emphasized, "To operate an actual CBDC system based on distributed ledger technology, it is necessary to establish a control system and develop related operational processes."

The Bank of Korea plans to continue linkage experiments this year by expanding the number of participating institutions.

This content was produced with the assistance of AI translation services.