Improved Estimation Services Boost DApp Time-Expense Balance by 255%

In the realm of decentralized applications (DApps), balancing transaction speed and fees is a critical challenge. Organizations developing DApps aim to provide a high quality of service (QoS) where transactions are processed within a specified time frame, such as 5 minutes, while minimizing expenses. This balance, referred to as the time-expense balance, is crucial for optimizing user experience and operational costs.

To achieve this balance, developers rely on estimation services that predict transaction processing times based on gas prices. These services provide lookup tables that help developers choose the optimal gas price for their transactions. For instance, a developer named Charlie might use an estimation service to determine the gas price that ensures a transaction is processed within 5 minutes at the lowest possible cost. However, the accuracy of these estimation services is a significant challenge, as they often underestimate or overestimate processing times.

In practice, Charlie might use an estimation service that is not perfect and always picks the third smallest price that satisfies the processing time criterion. This approach ensures that the transactions are processed within the desired time frame while staying within the budget. The effectiveness of the estimation service is evaluated using the harmonic mean of two variables: the percentage of transactions processed within the specified time and the remaining budget as a percentage. This method is analogous to measuring the balance between precision and recall through the F1-measure.

The accuracy of the estimation service directly impacts the time-expense balance. For example, if the estimation service is improved to be fifty percent more accurate, the percentage of transactions processed within 5 minutes and the remaining budget would both increase significantly. This improvement would lead to a better time-expense balance, ultimately benefiting the end-users of DApps. The results of this simulation show that a more accurate estimation service can lead to a 2.55 times better time-expense balance, highlighting the importance of accurate processing time estimation services.

The need for reducing transaction processing times is particularly important for mission-critical DApps, but any DApp that issues a reasonable amount of daily transactions would benefit from a better time-expense balance. For instance, a DApp with a QoS requirement of having 90% of transactions processed within 30 minutes could reduce its operational costs by half if developers can achieve the same QoS while spending less on gas prices. This underscores the key challenge for DApp developers: optimizing the time-expense balance to achieve the desired QoS at the lowest possible cost.

To address this challenge, accurate processing time estimation services are essential. These services help developers make informed decisions about gas prices, ensuring that transactions are processed within the desired time frame while minimizing expenses. The accuracy of these services is crucial for achieving a better time-expense balance, which ultimately benefits the end-users of DApps. In this paper, the focus is on determining how long transactions normally take to be processed in Ethereum and evaluating the accuracy of current existing services. This research aims to provide insights into the time-expense balance and the importance of accurate processing time estimation services for DApp developers.