https://doi.or g/10.31449/inf.v47i2.4965 Informatica 47 (2023) 297–298 297
T radeoffs In Using Blockchain T echnology For Security , Privacy , And
Decentralization:theor etical And Empirical Perspectives
Aleksandar T ošić
Faculty of Mathematics, Natural Sciences and Information T echnologies
University of Primorska
E-mail: aleksandar .tosic@upr .si
Thesis Summary
Keywords: Edge Computing, Blockchain, Container orchestration, Consensus mechanism
Received: June, 15, 2023
This paper is and extended abstract of the doctoral thesis [1]. It identifies four selected topics in which
blockchain technology can have a positive or transformative effect on existing solutions. W e pr opose new
pr otocols, which change the curr ent standar ds to add functionality , impr ove performance or over come lim-
itations of existing blockchain networks. Specifically , we focus on container or chestration on the edge
using a unique blockchain pr otocol for security , verifiability , and trust.
Povzetek: Pričujoče delo je razširjen povzetek doktorske disertacije [1]. Delo pr edstavlja raziskavo, ki se
osr edotoča na uporabo tehnologije blockchain za izboljšanje obstoječih r ešitev na štirih izbranih podr očjih.
Poseben poudar ek je na razvoju in implementaciji edinstvenega blockchain pr otokola za orkestracijo kon-
tejnerjev na r obu omr ežja, ki zagotavlja varnost, pr everljivost in zaupanje.
1 Intr oduction and pr oblem
statement
In recent years, cloud computing became a commonly used
architecture for most applications. The shift of the geogra-
phy of computation was incentivized by many factors rang-
ing from ease of software maintenance [2], reliable qual-
ity of service(QoS), hardware flexibility , and cost (CapEx
to OpEx) [3], etc However , with the expected growth of
data generation and consumption and storage and service
provisioning in cloud computing environments, the archi-
tecture is pushing network bandwidth requirements to the
limit [4]. Edge computing in it’ s simplest form can be de-
fined as an architecture in which computation is moved to
the edge of the network in order to make use of the geo-
graphic proximity to decrease latency and improve band-
width. This recent paradigm shift attempts to address
the overly geographically-centralized cloud architecture.
However , distributing services to the edge introduces new
challenges such as resource allocation, service and applica-
tion migration, trust, etc.. Blockchain technology may be
used to address some of the issues. It can serve as a layer
of trust between the system, and the end user by providing
a verifiable and transparent ledger of the state of the sys-
tem. T o achieve this, a new protocol is required that would
overcome the latency constraint, decentralized resource al-
location, and real-time container migrations [5].
2 Methodology
W e design, and develop a new blockchain protocol aimed
at autonomous decentralized container orchestration suit-
able for edge devices. The proposed protocol uses veri-
fiable delay functions(VDFs)[6] as the entropy source for
secure randomness. Nodes participating in consensus com-
pute a function p = vdf(bh,bd ) where p is the proof, bh is
the SHA256 hash of the current block, and bd is the dif fi-
culty of the current block. W e show that p is a suf ficiently
secure source of entropy for generating randomness. More-
over , given delay imposed on the nodes computing the VDF
prevents malicious nodes to peek into the future. Using the
shared seed, nodes are able to self-elect into consensus roles
for each slot without communication overhead as shown in
Figure 1.
3 Evaluation methodology and
r esults
W e performed extensive testing of our reference implemen-
tation simulating networks as lar ge as 1000 nodes. The
telemetry obtained from logging the state of all nodes shows
that the protocol is scalable, and can ef ficiently conver ge to-
wards average resource utilization of the entire network by
performing real-time container migrations between nodes
using Checkpoint/Restore In Userspace (CRIU). Moreover ,
our results show that using CRIU significantly improves the
performance making our protocol viable in practice.

298 Informatica 47 (2023) 297–298 A. T osić
V
V
V
V
V
V
V
V
V
V
V
V
V
R
V
R
V
R
V
P
shuffle(proof)
Slot n
Slot n +1
Block (n)
proof = vdf(n.blockHash, difficulty)
C
C
C
C
V
V
V
V
V
V
V
P
C
C
C
C
P C
C
C
C
P C
C
C
C
Self-elect into roles / Compute
clustering overlay
Gossip new block
T=0 T= 1/3 * slotTIme T= slotTIme T= 2/3 * slotTIme
Gather statistics and attestations
P R
R
R
R
Report Statistics
Candidate Block
Send Attestations
1
2
3
Execute 
Migrations
Figure 1: Role based consensus mechanism using VDFs as an entropy source
4 Discussion and further work
Our results showcase the feasibility of the proposed proto-
col for lar ge networks of edge devices with limited com-
pute resources. However , decentralized networks must ad-
dress Byzantine behaviour of nodes. T o secure the proto-
col against malicious actors the protocol must secure con-
tainerized application and guarantee the execution. Exist-
ing solutions such as Intel SGX are not generic and inher -
ently impose hardware restrictions on the protocol. Re-
search should focus overcoming specific hardware imple-
mentations of trusted computation.
Refer ences
[1] A. T ošić, “Empirična študija uporabe tehnologije ver -
iženja blokov v obstoječih sistemih in arhitekturah:
doktorska disertacija,” Ph.D. dissertation, Univerza na
Primorskem, Fakulteta za matematiko, naravoslovje
in …, 2022.
[2] E. Bayrak, J. Conley , and S. W ilkie, “The economics
of cloud computing,” 201 1.
[3] M. Armbrust, A. Fox, R. Grif fith, A. D. Joseph,
R. Katz, A. Konwinski, G. Lee, D. Patterson,
A. Rabkin, I. Stoica et al. , “A view of cloud comput-
ing,” Communications of the ACM , vol. 53, no. 4, pp.
50–58, 2010.
[4] W . Shi, J. Cao, Q. Zhang, Y . Li, and L. Xu, “Edge
computing: V ision and challenges,” IEEE Internet of
Things Journal , vol. 3, no. 5, pp. 637–646, 2016.
[5] A. T ošić, J. V ičič, M. Burnard, and M. Mrissa, “A
blockchain protocol for real-time application migra-
tion on the edge,” Sensors , vol. 23, no. 9, 2023.
[Online]. A vailable: https://www .mdpi.com/1424-
8220/23/9/4448
[6] D. Boneh, J. Bonneau, B. Bünz, and B. Fisch, “V er -
ifiable delay functions,” in Advances in Cryptology–
CR YPT O 2018: 38th Annual International Cryptology
Confer ence, Santa Barbara, CA, USA, August 19–23,
2018, Pr oceedings, Part I . Springer , 2018, pp. 757–
788.