Abstract

Edge and Internet of Things (IoT) applications have attracted significant attention from both industry and academia due to their immense potential. As a result, the database community—through communications such as the recent database Seattle reports—has recognized the criticality of developing a new breed of data management systems specifically tailored for IoT and edge applications. These systems need to be distributed across edge locations to effectively handle the unique challenges posed by these environments. However, the development of such databases remains largely minimal in both industry and academia. Over the past five years, our team has conducted extensive research and collaborated with industry partners to bring an edgecloud database to market and to investigate the reasons behind the limited progress and adoption of distributed edge-cloud databases. As part of our efforts, we have developed a distributed edge-cloud database called AnyLog and deployed it with numerous industry partners in different IoT and smart city sectors. Our interactions with many technical teams from diverse industries have provided invaluable insights into the challenges and opportunities associated with distributed edge-cloud data management. In this paper, we present a comprehensive summary of our findings, drawing from our five-year experience in the field. We highlight the real challenges faced in distributed edge-cloud data management and discuss the opportunities that lie within. Furthermore, we showcase the capabilities of edge-cloud databases using AnyLog, illustrating our insights about the challenges and opportunities of edge-cloud databases.