Kadabra: Adapting Kademlia for the Decentralized Web (2210.12858v2)

Abstract: Blockchains have become the catalyst for a growing movement to create a more decentralized Internet. A fundamental operation of applications in a decentralized Internet is data storage and retrieval. As today's blockchains are limited in their storage functionalities, in recent years a number of peer-to-peer data storage networks have emerged based on the Kademlia distributed hash table protocol. However, existing Kademlia implementations are not efficient enough to support fast data storage and retrieval operations necessary for (decentralized) Web applications. In this paper, we present Kadabra, a decentralized protocol for computing the routing table entries in Kademlia to accelerate lookups. Kadabra is motivated by the multi-armed bandit problem, and can automatically adapt to heterogeneity and dynamism in the network. Experimental results show Kadabra achieving between 15-50% lower lookup latencies compared to state-of-the-art baselines.

• The paper introduces Kadabra, which optimizes Kademlia for faster data lookups in decentralized networks using innovative adaptive routing table computations.
• It leverages principles from the multi-armed bandit problem to balance exploration and exploitation in managing network heterogeneity.
• Experiments demonstrate a 15–50% reduction in lookup latency, highlighting Kadabra's potential to bolster performance in decentralized web applications.

The paper "Kadabra: Adapting Kademlia for the Decentralized Web" addresses the need for efficient data storage and retrieval in decentralized Internet applications. With blockchains at the forefront of creating a decentralized web, current blockchain technologies face limitations regarding storage functionalities. As a solution, various peer-to-peer data storage networks have been developed, often leveraging the Kademlia distributed hash table (DHT) protocol. However, the authors argue that existing implementations of Kademlia are insufficient for the rapid data operations required by decentralized web applications.

Key Contributions

1. Kadabra Protocol: The paper introduces Kadabra, an enhanced decentralized protocol designed to improve Kademlia's efficiency. By optimizing the computation of routing table entries, Kadabra aims to accelerate data lookups within the network.
2. Inspiration from the Multi-Armed Bandit Problem: Kadabra is motivated by concepts from the multi-armed bandit problem, an optimization issue where one must balance exploration and exploitation to find optimal solutions. This approach allows Kadabra to adaptively manage network heterogeneity and dynamism.
3. Performance Improvements: Experimental evaluations indicate that Kadabra achieves significantly lower lookup latencies, ranging from 15-50%, in comparison to current state-of-the-art protocols. This improvement suggests that Kadabra could effectively support the needs of decentralized web applications, which rely on fast data transactions.

Significance

The research aims to overcome specific inefficiencies in Kademlia, which is widely utilized in peer-to-peer networks for decentralized applications. By employing adaptive techniques from the multi-armed bandit problem, Kadabra shows potential in enhancing network performance, making it a promising solution for facilitating more efficient decentralized Internet operations.

Ultimately, Kadabra contributes to the ongoing efforts to build decentralized web infrastructures that can match the performance demands of modern applications, thereby advancing the utility and scalability of decentralized networks.