Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Blockchain Access control Ecosystem for Big Data Security

Blockchain Access control Ecosystem for Big Data Security

Presented at the IEEE International Conference on Blockchain, this talk discusses how blockchain access control can be used to secure big data. The case study was for securing plant genotype data.

Uchi Uchibeke

July 31, 2018
Tweet

More Decks by Uchi Uchibeke

Other Decks in Science

Transcript

  1. Blockchain Access control Ecosystem for Big Data Security IEEE International

    Conference on Blockchain Uchi Uchibeke Sara Hosseinzadeh Kassani Kevin A. Schneider Ralph Deters July 31, 2018 1
  2. Key Challenges 1. Security Security issues in processing, storing, and

    transferring of confidential or sensitive data are a serious concern that could cause ethical, intellectual property and privacy concerns. 3
  3. Key Challenges 2. Access Control Management Data misuse or theft

    can cost an organization millions of dollars. 4
  4. Key Challenges 3. Authentication and Authorization: Authentication and Authorization is

    another key issue and it should ensures only specific individuals have access to data, making the data unavailable to all others. 5
  5. Key Challenges 4. Data Ownership Big Data storage and processing

    services is concerned with the issues of data sharing in cloud environments due to untrustworthy, multi-ownership and dynamic nature of these environments. Providing proper data ownership roles support integrity and confidentiality of data and privacy of participants from the third party. 6
  6. Key Challenges 5. Intellectual Property Intellectual property rights (IPR) management

    is an important part of any data management program. • Some data by nature is sensitive, confidential, or both. • The owners of data resource will have an interest how others may use it. • The IP Rights ensures all legal, ethical, and professional obligations of the data are met by researchers. 7
  7. Process What was done § Model a Business network using

    Hyperledger Composer § Create assets and participants § Interact with Business network from web app and via SMS 10
  8. Hyperledger Fabric and Composer § Rapid prototyping and testing with

    Hyperledger Fabric § Hyperledger Fabric is private and permissioned § Hyperledger assumes an environment of partial trust 11
  9. Solutions § Two solutions a) Blockchain Identity Based Access Control

    (BIBAC) b) Blockchain Role Based Access Control (BRBAC) § Based on existing access control paradigm § Both solutions were compared 12
  10. Results and Lessons learned § Flexibility § Modularity § Connection

    to any HTTPS enabled device § BRBAC seems more suitable for entreprises 15
  11. Blockchain vs. Database APPENDIX 1 Blockchains are all about TRUST

    With a database, all parties have to have full trust for the holder of the database. Where as on a Blockchain, the trust between the bank employee and the client is enforced by the Blockchain. Only the client can unlock the sign up process using their secure private key. This means that no nefarious employees can sign customers up for products and services without client consent. “5,300 Wells Fargo employees fired over 2 million phony accounts” -CNN MONEY, September 2016 Events like what happened at Wells Fargo have eroded client trust in banking. Blockchain technology can re-establish client trust by ensuring nothing like this could happen ever again.
  12. Why use a Blockchain? APPENDIX 2 Blockchains provide many more

    benefits over traditional database systems. BLOCKCHAINS: • No Single Point of Failure • Are resistant to ransomware hacks like WannaCry • Data is highly available because it is distributed • With every block creation, a new layer of encryption is added to secure past transactions • We can create transactions to provide full analytics on the data • Smart Contracts can automate work flows right on the chain
  13. How many nodes are in your blockchain? APPENDIX 3 For

    development purposes, we built our blockchain to have 4 nodes. These nodes run on Docker instances on our Amazon Web Services Development Environment. In production, the more nodes we can establish throughout the organization, the more secure our network can be. We can have different types of nodes. • Full nodes that hold all of the data and perform all of the tasks of a node. • Light nodes that interface with the Blockchain but don’t strain the system that they run on.
  14. Why did you pick IBM’s Hyperledger over other Blockchains like

    Ethereum APPENDIX 5 Hyperledger was designed with private Blockchains in mind for use in institutions like banks. Hyperledger does not expose the client data in full to each and every node and the blockchain cannot be queried by anybody as is the case with Ethereum. Because Ethereum is such a vast network of nodes, to achieve consensus can take a long time. On Hyperledger, private blockchains achieve consensus in much less time using a specialized consensus algorithm giving us much better performance. VS.