Atlas Project: Mapping the world of decentralized finance

Author: BIS Compiler: Hong Caixuan

Recently, the Bank for International Settlements (BIS) issued a document detailing the Atlas (Project Atlas: Mapping the world of decentralised finance). The project aims to provide central banks and financial regulators with data on crypto assets and decentralized financial markets. Crypto assets and decentralized finance (Defi) applications are part of a global emerging financial ecosystem, but these markets often lack transparency and can pose risks to financial stability. Therefore, the Atlas project provides central banks and financial regulators with tailored data by integrating data from crypto transactions and data from public blockchains to more accurately assess the economic significance of these markets. Its self-developed platform and infrastructure provide central banks with technical and analytical capabilities and publicly share data and results. The Institute of Financial Science and Technology of Chinese Minmin University compiled the core part of the research.

One

Introduction

Crypto assets and decentralized finance (DeFi) applications are part of an emerging financial ecosystem that spans the globe. Since Bitcoin’s debut in 2009, this ecosystem has seen phenomenal growth. Today, there are thousands of crypto assets with a total market capitalization of more than $1 trillion. Due to its large number of participants, protocols, and networks, the ecosystem formed by cryptocurrencies and decentralized finance is very complex and difficult to analyze.

While introducing new technologies, the crypto market tends to lack transparency and present risks. Regulators are increasingly concerned about the size of the crypto market and its integration with traditional finance. The cryptocurrency and DeFi markets have left a considerable data gap. For example, information on cross-border capital flows in macrofinance is difficult to obtain. The Atlas project attempts to answer the question: How can central banks use technology to gain insight into decentralized financial markets and gather more reliable and comprehensive statistics on the crypto ecosystem?

The Atlas project aims to bridge some of the existing data gaps by developing a specialized data platform with a variety of data sources to shed light on the macroeconomic implications of cryptocurrencies and DeFi. **Proof of Concept (PoC) provides insights into cross-border flows. According to the needs of central banks and financial regulators, it fuses data collected from crypto exchanges (off-chain data) with public blockchain data collected through operational nodes (on-chain data), with the benefit of creating new statistics and reviewing existing data.

**As part of the first PoC, the Atlas project focuses on the flow of Bitcoin between crypto exchanges across geographic locations. Atlas integrates diverse data sources and builds strong technical and analytical capabilities to support the central banking community, providing policymakers, analysts and researchers with granular data and customized metrics.

Two

The need for tailored and reliable data in the crypto market

Despite the plethora of data in the cryptocurrency and DeFi markets, the reasons why central banks adopt custom platforms are as follows:

1. Encrypted data is generally not tailored to the needs of central banks. **There is currently a lack of data tailored to central banks and financial regulators to support the analysis of the macroeconomic correlation of the crypto market and the potential impact on financial stability.

**2.Collecting cryptocurrency and DeFi data differs from data collection practices common in traditional financial systems. **Atlas transforms and stores encrypted data on a scalable cloud platform, providing a starting point for comprehensive data analysis based on a transparent approach.

**3.The analysis of the crypto market needs to take into account its unique characteristics. **Atlas helps build technical and analytical capabilities that enable central banks to respond to new developments in highly dynamic markets.

4.Although blockchain data aggregation methods tend to be a black box, this prevents researchers and regulators from benchmarking data from different sources. Using fine-grained data and combining different data sources, Atlas can help identify inconsistencies between data sources and why they can occur.

**5.Data is prone to manipulation and false positives. **Atlask can review data and calculate statistics based on granular data. Microdata supports sensitivity analysis and a clear understanding of data sources to help inform regulatory initiatives.

The following figure explains the application scenarios of Atlas according to the target audience and analysis class. Atlas can provide researchers with microdata to study economic developments over time, or to analyze specific events in particular. At the same time, the platform can serve policymakers through indicators that facilitate ad hoc analysis and assessment of long-term developments.

Figure 1 Application scenarios of Atlas projects among researchers and policymakers

Three

Project Overview

**Atlas projects consist of data ingestion, data platforms, and visualizations in dashboards. The data platform and pipeline are at the heart of the project. The pipeline transfers (granular) raw data to the analytics platform and generates images using the dashboard (Figure 2). The platform is set up with scalability in mind. It is implemented as a cloud sandbox with the ability to adjust IT resources as needed, allowing users to access outputs in an analytics environment. The specific description of each part is as follows:

Figure 2 Atlas project overview

1. Data ingestion and processing: The Atlas platform builds and maintains data pipelines to ingest and transform on-chain and off-chain data into databases hosted in a sandbox in the cloud (Figure 3). The configuration and selection of the platform supports loading data from any source in its native data format: large amounts of on-chain data and granular or aggregated off-chain data. The platform’s processing logic integrates multiple data sources, which are enhanced and fused together to enable adaptive data analysis.

Figure 3 Illustration of data ingestion and processing process

2. Cloud infrastructure: Implemented on the Microsoft Azure platform, leveraging widely available or easily replaceable components. ETL (Extract, Transform, Load) pipelines transform and process incoming raw data into common data structures and load it into a database.

3. Analytics Environment: An analytics environment provides advanced users with familiar tools and an extensible platform. Through the interactive coding notebook interface provided by the platform, users can access data stored in different layers. The data description includes the data source, the data context (such as Bitcoin stream or entity master data), variable names and formats, and sample data. Users have the flexibility to combine data sources according to their needs and take advantage of advanced analytics within the platform. In addition, the platform facilitates the creation of visualizations and the export of aggregated data to the local environment.

4. Dashboard: Pre-calculated metrics serve a front-end web application that presents intuitive and interactive dashboards that provide users with a comprehensive view of key metrics and trends, as well as drill-through for data exploration. (More on that below)

5. Processing on-chain data: To collect on-chain data from the Bitcoin network, Atlas runs a dedicated blockchain node. Upon reaching the copper layer (Figure 3), the Bitcoin data is transformed into a structured table that includes information such as the inputs and outputs of transactions. For the silver tier, Atlas further converts the data into a format familiar to many analysts, similar to payment system data, with sender, receiver, transaction value and timestamp, as well as more detailed information about the transaction. For the gold tier, the processing engine aggregates the data in different ways, such as the total transactions per day. Off-chain data directly or indirectly complements on-chain data.

Four

First Proof of Data Sources and Concepts

On data sources, Atlas uses a big data approach to integrate high-volume and high-velocity data in various formats. Data comes from two sources: on-chain, from the public ledger; Off-chain data, reported by market participants, or more generally, any data other than data extracted from the ledger.

**On the proof of concept simulating cross-border flows, as shown in the figure below, the Atlas project data uses transactions between crypto exchanges in the Bitcoin network, as well as the locations of these exchanges, as proxies for cross-border capital flows. **The jurisdiction of crypto exchanges is not always identifiable due to multiple locations or the lack of a central location for registration, so estimates of cross-border traffic should be considered as a floor estimate. In future iterations, the affiliation of other countries will be explored. This process may include mapping exchanges to countries based on user distribution.

Figure 4 Cross-border traffic is derived based on cryptocurrency exchange locations

Four

First Proof of Data Sources and Concepts

The Atlas project’s dashboard visualizes data aggregation and analysis results, providing correlation insights into market players at the heart of the crypto ecosystem**. **On-chain transaction categories show different segments of on-chain traffic in the form of time series.

**Drill down into the total transactions, the value interval in which the transaction is located can indicate the correlation of on-chain crypto transactions with retail payments and remittances. **This type of payment usually falls within a smaller range of value. However, further criteria are needed to reliably determine the purpose of such transfers. The chart below distinguishes Bitcoin transactions valued under $500 and under $200. The value of these transactions is almost negligible compared to the volume, indicating that on-chain transactions are highly skewed towards large transfers. However, when looking at the number of transactions, they make up the majority of transfers.

Figure 5 Visualization results of transaction value range

Visualizing data can also provide a global overview of traffic – how they have evolved, as well as their size and concentration. **Policymakers and regulators can amplify traffic and gain new insights relevant to their jurisdiction.

Figure 6 Globalization of cross-border flows

In addition, the Atlas project can use scatterplots to visualize the relationship between on-chain and off-chain activity. For example, as a large number of users trade on cryptocurrency exchanges, exchanges should show higher token holdings and increased on-chain traffic. The chart below plots on-chain inflows and trading volumes for each year. Taking into account the highly skewed value, the data is displayed on a logarithmic scale. One data point in the graph represents an exchange and its associated on-chain traffic and reported volume for a given year.

Figure 7 Scatter plot, transaction volume and on-chain traffic

Five

Conclusion and Outlook

Faced with the complex cryptocurrency and decentralized financial markets, Atlas serves researchers and analysts, as well as policymakers, supporting structural analysis and providing user-friendly tools to generate intuitive insights. As a starting point for the initial assessment, the Atlas project provides information for the drafting of data reporting requirements and regulation of crypto market participants. Based on a clear and transparent approach, Atlas improves the technical capabilities of central banks and enhances the understanding of cryptocurrency and DeFi data. **

**Atlas provides a starting point for regulators and the central banking community to build a comprehensive crypto-asset analytics environment that can be extended to new data sources and analytical approaches. Today, in an effort to increase transparency, several centralized crypto exchanges expose information on the addresses of crypto assets they control — under the label of “proof of reserve.” In the future, this disclosure could be used for new approaches to data-driven regulation.

**Atlas’ next important task is to include more data. **As the DeFi market expands, it will be possible to extract and analyze data from blockchain nodes in Ethernet in the future. In addition, the project team will seek feedback from the central banking community on how statistical data can help them fulfil their mandate. The data platform and dashboard of the Atlas project will be made available to test users in order to gather feedback on functional and data needs. This will inform the development of additional features and visualizations.