Skip to content. | Skip to navigation

Informatik 5
Information Systems
Prof. Dr. M. Jarke
Sections
Personal tools
You are here: Home Theses Analyzing distributed ledger data structures for the immutable persistence of transactions and tamper-secure execution of processes

Contact

Prof. Dr. M. Jarke
RWTH Aachen
Informatik 5
Ahornstr. 55
D-52056 Aachen
Tel +49/241/8021501
Fax +49/241/8022321

How to find us

Annual Reports

Disclaimer

Webmaster

 

 

Analyzing distributed ledger data structures for the immutable persistence of transactions and tamper-secure execution of processes

Thesis type
  • Master
Status Open
Supervisor(s)
Advisor(s)
  • Thomas Osterland

Often only noticed as a technology that enables the digital currency Bitcoin, blockchain is a novel protocol that allows the distributed and secure storing of information and untampered execution of program code in trust-less environments. As part of a broader range of technologies that allow the immutable storage of information in decentralized networks, where network participants build a consensus about the correctness of persisted information, the classical blockchain is only one specific implementation of distributed ledger technologies. Other ledger technologies are based on directed acyclic graphs (DAGs). Topic of this thesis is the exploration and implementation of a DAG based ledger in the existing blockchain framework Labchain (https://github.com/putschli/LabChain).

 

The whole history of blockchain started 2008 with the popular paper of Satoshi Nakamoto https://bitcoin.org/bitcoin.pdf. He presented an approach that enabled the collaborative handling of information such that no single entity can tamper with persisted data. This concept was extended to secure the execution of program logic with the introduction of smart contract enabled blockchains. One important tool to guarantee this characteristic is the blockchain data structure itself, that aggregates transactions in blocks and contain the hash of the preceding block. In this way it is not possible to remove or change a block within the chain without changing every subsequent block. Together with the consensus that is build between the network participants it is not possible for a single entity to change the information stored in the blockchain.

Although the blockchain data structure is the most popular and broadly applied in many different technologies, it has some drawbacks. So it is only possible to aggregate a limited number of transactions in one block, because otherwise the network latency for transmitting this block becomes a problem for the consensus stability, that means new blocks are not efficiently communicated throughout the network what leads to increased branching and thus affects the overall ledger performance. Other drawbacks covering the efficiency for finding consensus among network participants of block based distributed ledgers or the time one need to wait until a transaction is securely added to the blockchain and thus tamper-secure. (An overview of existing problems is i.a. given here: http://ceur-ws.org/Vol-1938/paper-kot.pdf)

Other data structures such as trees or directed acyclic graphs (DAGs) are proposed for storing transactions and providing equal functionality regarding the maintenance of consistency as the traditional chain-oriented blockchain data structure. There exists approaches, as Phantom https://eprint.iacr.org/2018/104.pdf or Spectre https://eprint.iacr.org/2016/1159.pdf which propose the introduction of a DAG data structure on a block level and others, as for instance the popular IOTA https://assets.ctfassets.net/r1dr6vzfxhev/2t4uxvsIqk0EUau6g2sw0g/45eae33637ca92f85dd9f4a3a218e1ec/iota1_4_3.pdf that use a DAG on a transaction level. Other such competitors, are i.a. Byteball https://obyte.org/Byteball.pdf, Swirdls https://www.swirlds.com/downloads/SWIRLDS-TR-2016-01.pdf, Raiblocks https://raiblocks.net/media/RaiBlocks_Whitepaper__English.pdf and many more.

Often DAG based ledger systems have stability issues, as for instance DAG splitting, where the DAG that should continuously grow in one direction is split in two different parts, that will never be merged again, or the problem that certain transactions will never become confirmed by other succeeding transactions. Studies for IOTA that simulate and benchmark the stability of the tangle are provided here https://hal.archives-ouvertes.fr/hal-01716111v2/document and here https://assets.ctfassets.net/r1dr6vzfxhev/2ZO5XxwehymSMsgusUE6YG/f15f4571500a64b7741963df5312c7e7/The_First_Glance_of_the_Simulation_Tangle_-_Discrete_Model_v0.1.pdf

Other technologies as Byteball follow different approaches claiming that those solve the problems of IOTA.

For these reasons, we are interested in an analysis of different data structures and a thorough benchmark comparing advantages and disadvantages of different alternatives. Part of the thesis is the implementation of a DAG data structure and the necessary consensus into the existing LabChain (https://github.com/putschli/LabChain) framework. The aim is to compare different existing technologies and to provide experimental results.

 

Document Actions