DataWalk Performance Test for Multi-Node Scaling

This technical report presents the results of a performance test conducted on the DataWalk software platform, specifically evaluating its scalability when executing the Weakly Connected Components (WCC) graph algorithm using data from the Bitcoin blockchain. The testing involved running the WCC algorithm against a dataset comprising over three billion nodes and edges, demonstrating that DataWalk can effectively scale linearly across six compute servers. The report details the performance metrics for both data indexing and WCC execution times, showing significant improvements in processing speed as additional compute nodes were added. For instance, the data indexing time decreased from 91 minutes with one server to just 22.5 minutes with six servers, while WCC execution time improved from 10 minutes to 2.8 minutes under the same conditions. The findings indicate that DataWalk is capable of handling large volumes of data efficiently, making it a viable solution for complex data analytics tasks.