The global banking cooperative SWIFT has released an analysis of its recent proof of concept that relied on IBM’s Hyperledger Fabric v1.0 DLT platform. While it’s too soon for the cooperative’s entire network of banks to adopt a DLT solution, SWIFT representatives appeared extremely pleased with the results of the sandbox trial.
On March 8, the financial cooperative SWIFT, the member banks of which number around 11,000, issued a press release[1] proclaiming the success of its recent attempt to use Hyperledger's[2] distributed ledger technology (DLT) to "support automated real-time liquidity monitoring and reconciliation."
The proof of concept (PoC), which SWIFT calls "one of the largest and most ambitious" DLT trials," aimed to test the efficacy of permissioned distributed ledgers in reconciling Nostro[3] accounts, which are bank accounts belonging to one bank that exist on a different bank's books. This involves two banks that are located in different countries and therefore handle different currencies.
The sandbox project explored whether DLT, together with SWIFT assets, "would meet industry-level governance, security and data privacy requirements." It also sought to determine how close DLT has come to being able to service SWIFT members.
According to the press release:
"[The technology] enabled real-time event handling, transaction status updates, full audit trails, visibility of expected and available balances, real-time simplified account entries confirmation, the identification of pending entries and potential related issues, and generated the data required to support regulatory reporting. The PoC also demonstrated the significant progress DLT has made with regards to data confidentiality, governance, security, and identification frameworks."
An analysis [4]of the PoC, also released by the cooperative, relates that the consensus process requires both parties involved in a transaction to approve it, and that SWIFT must also "check" each transaction.
34 banks took part in the PoC, each running its own node, and 528 channels had to be created in order to ensure that certain private data were not shared to the entire sandbox. In order to make the solution capable of servicing the SWIFT network in full, some 100,000 such channels "would need to be established and maintained, covering all existing Nostro relationships, presenting significant operational challenges."
Another obstacle to scaling is the need for participants in a DLT-SWIFT system to "migrate from batch to real-time liquidity reporting and processing, [while] back office applications would need to be upgraded to feed the platform with real-time updates."
All the same, the press release suggests that together, DLT and SWIFT assets can provide "the necessary foundation for financial multi-bank applications." The cooperative pledged to work with its members to overcome some of the roadblocks inhibiting network-wide DLT adoption.
Overall, the "PoC went extremely well, proving the fantastic progress that has been made with DLT and the Hyperledger fabric in particular," rhapsodized SWIFT head of research and development Damien Vanderveken.
Chief Platform Officer Stephen Gilderdale stated that SWIFT had already embarked on other PoCs, and that
