Blockchain Enabled Distributed Storage and Sharing of Personal Data Assets, presentatio by Jānis Grabis, Vlado Stankovski, and Roberts Zariņš at BlockDM 2020

1. 1
Jānis Grabis
Riga Technical University
Riga, Latvia
grabis@rtu.lv
Vlado Stankovski
University of Ljubljana
Ljubljana, Slovenia
Vlado.Stankovski@fri.uni-lj.si
Roberts Zariņš
Riga Technical University
Riga, Latvia
roberts.zarins@gmail.com
Blockchain Enabled
Distributed Storage and
Sharing of Personal Data
Assets

2. 22
 Data sharing has potential of creating value
to their consumers as well as their owners
 Centralized management
– Limited control and value-creation opportunities
 Decentralized management
– No single point of control
– Blockchains and smart contracts enable data and
value exchange
– Limitations concerning both storage volume and
information processing speed
Background

3. 33
 Zyskin et al. (2015) propose using
blockchains for personal data management
 Singh and Lee (2018), Kosteka et al. (2017)
and Pongnumkul et al. (2017) analyze QoS
concerns of using blockchains for personal
data management
 Zheng et al. (2015) and Bartoletti et al. (2017)
focus on querying languages for analyzing
cryptocurrency transactions.
 Alessi et al. (2019) and Wang et al. (2018)
combine on-chain and off-chain processing
Existing Work

4. 44
Objective
• Blockchain controlled access to distributed personal data
according to conditions specified in smart contracts
• Knowledge base as a mediator for efficient querying of
distributed personal data to meet high Quality of Service
requirements
To elaborate a method for efficient
distributed storage and sharing of personal
data assets within a community of users

5. 55
MyDataExperience: Components of
Personal Data Management Solution

6. 66
Data Management Processes
sd Search
Distributed
storage
BlockchainKnowledge base
User
Search assets()
Pre-select assets()
Validate access()
Provide access()
Search results as meta-data()
Request selected assets()
Request assets()
Return results()
Send
assets()
• Store personal data
assets in distributed
storage
• Share personal
data assets taking
into account privacy
preferences
• Search of
personal assets
shared by related
entities

7. 77
 React framework based front-
end
 Ethereum blockchain
– Ropsten test network
– Infura API
 Transactions are handled
using Ethereum Wallet
 Smart contracts were
developed using the Solidity
programming language
 InterPlanetary File System
(IPFS) provides distributed
storage facilities
Technology

8. 88
 Comparison
– MyDataExperience
– On-chain storage of data assets
 Performance measures
– Storage size
– Data assets search efficiency.
Experimental

9. 9
Storage Size
File
number
Hex file
size, b
Total size, b Gas, unit Cost,
Eth
BC size increase,
b
On-chain
1 2048 2436 1622719 0,041 2436
2 6438 6852 4691848 0,117 9288
3 9440 9828 6762981 0,160 19116
MyDataExperience
1 46 356 215089 0,005 356
2 46 356 215089 0,005 712
3 46 356 215089 0,005 1068
• Three files are stored in
MyDataExperience and on-chain solution
• Measures
• Size of the file’s hexadecimal data
string
• Total size of data stored in the
blockchain
• Ethereum gas consumption
• Transaction costs

10. 10
File Search
Nr Scenario
1 Find all files owned by friends
2 Find all files owned by the user herself
3 Find all files shared with the user
4 Find all PDF files owned by friends
5 Find all files titled
“Blockchain_Whitepaper” and owned by
friends
6 Find all files owned by friend
0xd335B51d960061cC1FA145e6d4a1F
cf1aB55677C

11. 1111
 MyDataExperience solution is shown to adhere to the
principles Privacy-by-design
– Provides its users a full control on the way personal data assets
are stored and shared in the community of users
 The storage space is not dependent on the size of files stored
 Utilization of the knowledge base expedites the search
process
 Limitations
– Small-scale experimentation
– Cost of maintaining the knowledge base in the distributed mode
– Knowledge base itself could be redeveloped by using the
MyDataExperience approach by combining on-chain and off-
chain solutions
Conclusion

12. 12
grabis@rtu.lv
http://iti.rtu.lv/vitk/lv/katedra/darbinieki/janis-grabis
Thank you!