In this thesis, we aim to improve the performance of existing blockchains by removing common bottlenecks in system design. To this end, we study common bottlenecks in the system design of existing blockchains and propose mechanisms to mitigate them. Our contribution is twofold: first, we designed ALDER, a generic construction to improve the performance of leader-based blockchains. ALDER extends the consensus algorithm of existing leader-based blockchains to be enriched with multiple leaders cooperating to handle user requests. ALDER's design increases the throughput and latency of existing blockchains by eliminating consensus and network bottlenecks. We provide a detailed evaluation of ALDER through large-scale experiments. Second, we performed an in-depth analysis of IDA-Gossip: a blockchain-based gossip dissemination protocol. IDA-Gossip is an essential primitive for blockchains to eliminate the bottleneck of the store-and-forward gossip dissemination mechanism, but its properties have not been extensively studied. Our work quantifies the performance of IDA-Gossip with different protocol parameters under faults using experiments and simulations. Our investigation aims to help researchers understand the properties of IDA-Gossip by revealing its behavior under different conditions.