Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Master's Thesis |
| Scope | Discipline-based scholarship |
| Title | Design and Implementation of a Scalable IoT-based Blockchain |
| Organization Unit | |
| Authors |
|
| Supervisors |
|
| Language |
|
| Institution | University of Zurich |
| Faculty | Faculty of Business, Economics and Informatics |
| Number of Pages | 63 |
| Date | 2019 |
| Abstract Text | This master's thesis documents the design, implementation and evaluation of the sharding enabled Bazo blockchain. The decision to leverage sharding as the layer 1 scalability technique was the result of reviewing recent literature with respect to mainstream approaches to scale blockchains. The system supports sharded environments with one validator per shard. In case of multiple shards, the validators validate and store only a fraction of the whole incoming transaction load, and as such of the whole blockchain. Received transactions from the clients are assigned to shards based on their public address. The single shard chains synchronize the global state at every block height to make sure that the shards grow at the same pace. In order to do so, the new data type state transition was introduced which is emitted after every block creation from each shard. The state transition includes everything needed for the other shards to update their local state. Furthermore, the blockchain runs in xed lengths of X blocks called epochs. After each epoch, the validators are randomly re-assigned to the single shards. The system was evaluated through unit tests sending up to 950 transactions per second to the blockchain. The set-up with one validator and one shard achieved a peak transaction validation rate of 55,2 Tx/sec. In a similar set-up, Bazo was testet with two validators and two shards. Surprisingly, the overall TPS value decreased slightly with a peak of 48,8 Tx/sec. In this case, the main performance bottleneck turned out to lie in the synchronisation overhead of the two shards. Furthermore, the tests were extended to incorporate three valudators and three shards. Hereby, it was observed that the TPS capacity increased to a maximum of 59,5 Tx/sec, demonstrating that sharding has a positive in uence on the scalability of Bazo. Although the current system exhibits evident limitations and downsides, it nevertheless showed to be a rst feasible step towards making Bazo more scalable while incorporate the promising technique of sharding. |
| PDF File |
Download
|
| Export |
BibTeX
|