Computer Performance Engineering

s of Invited Talks Cryptocurrency and Blockchain Technology: Challenges and Opportunities William J. Knottenbelt Imperial College Centre for Cryptocurrency Research and Engineering, Imperial College London, London, UK wjk@imperial.ac.uk The meteoric rise of blockchain-enabled cryptocurrencies, and Bitcoin [2] and Ethereum [1] in particular, has received global attention, not least from governments, entrepreneurs and researchers. Cryptocurrencies, of which there are now more than 800, provide an attractive alternative to traditional fiat currencies via a distributed, trustless and self-governing framework which not only enables low-friction financial transactions around the globe but also preserves the freedom and privacy of spending inherent in cash transactions. Cryptocurrency and blockchain technology brings with it a host of new challenges from the quantitative modelling perspective. Indeed, a range of issues including performance, security, energy use, incentives and scalability are poorly understood, as are the inherent trade offs between them, despite these being critical barriers to mass adoption. What analyses are carried out often do not take into account problems posed by the lack of diversity that emerges from a natural tendency towards dominant concentrations of computational and other power. These can arise from something as simple as the majority of network participants flocking to deploy the most energy-efficient cryptocurrency mining hardware. Indeed it is estimated that up to 70% of the computational power assuring the integrity of the Bitcoin network is provided by a single model of a hardware device. This device was recently found to have a backdoor that could be used by the manufacturer to shut the device down. This talk will cover some of the challenges and opportunities posed in this context, with a special emphasis on the performance evaluation and quantitative modelling perspectives. It turns out that classical performance evaluation techniques, especially Markovian analysis and queueing theory, are readily applicable to the study of cryptocurrencies and blockchains. Further, a judicious combination of analytical modelling, simulation and benchmarking techniques can be effectively applied to yield insights. Building on [3], we will illustrate this in the context of a study of a queue-based Ethereum mining pool [4] whose superficially fair reward scheme turns out not only to penalise more powerful miners, but also to incentivise a number of attacks which can 1 W.J. Knottenbelt—The content of the talk is the result of joint work with A. Zamyatin, K. Wolter, C. Mulligan, P. Harrison, S. Werner and I. Stewart, amongst others. 1 Source: http://coinmarketcap.com. Accessed 5 July 2017. 2 Source: http://antbleed.com. Accessed 5 July 2017. increase rewards, including the donation of mining power to other participants in certain circumstances. Examples of such attacks observed in the real world will be presented. The talk will conclude by outlining student-led spinout activity and ongoing directions of research in the Imperial College Centre for Cryptocurrency Research and Engineering. The former includes Gradbase, a qualification verification startup, Aventus, a blockchain-based ticketing company and Kotiva Technologies, who are seeking to use blockchain technology to increase the integrity of supply chains. The latter includes work being supported by industrial partners such as Blockchain.com and Outlier Ventures, as well as grants sponsored by government-related bodies such as Innovate UK.

[1]  T. Kurtz.  Strong approximation theorems for density dependent Markov chains , 1978 .

[2]  Rajeev Alur,et al.  A Theory of Timed Automata , 1994, Theor. Comput. Sci..

[3]  Paolo Ballarini,et al.  Analysing oscillatory trends of discrete-state stochastic processes through HASL statistical model checking , 2014, International Journal on Software Tools for Technology Transfer.

[4]  Elza Erkip,et al.  User cooperation diversity. Part I. System description , 2003, IEEE Trans. Commun..

[5]  Boudewijn R. Haverkort,et al.  A logic for model-checking mean-field models , 2013, 2013 43rd Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN).

[6]  Christos Langaris,et al.  Non-preemptive priorities and vacations in a multiclass retrial queueing system , 1996 .

[7]  T. Kurtz.  Limit theorems and diffusion approximations for density dependent Markov chains , 1976 .

[8]  Marco Beccuti,et al.  (Stochastic) Model Checking in GreatSPN , 2014, Petri Nets.

[9]  K. J. Ray Liu,et al.  Cognitive multiple access via cooperation: Protocol design and performance analysis , 2007, IEEE Transactions on Information Theory.

[10]  Luca Bortolussi,et al.  Fluid Model Checking of Timed Properties , 2015, FORMATS.

[11]  Evsey Morozov,et al.  Stability analysis of a multiclass retrial system with classical retrial policy , 2017, Perform. Evaluation.

[12]  Sandeep Krishna,et al.  A Wnt oscillator model for somitogenesis. , 2010, Biophysical journal.

[13]  Evsey Morozov.  The tightness in the ergodic analysis of regenerative queueing processes , 1997, Queueing Syst. Theory Appl..

[14]  Walter L. Smith,et al.  Regenerative stochastic processes , 1955, Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences.

[15]  Anthony Ephremides,et al.  Relay-Assisted Multiple Access With Full-Duplex Multi-Packet Reception , 2013, IEEE Transactions on Wireless Communications.

[16]  E. Morozov,et al.  Stability analysis of regenerative queueing systems , 2009 .

[17]  Marco Beccuti,et al.  Approximate analysis of biological systems by hybrid switching jump diffusion , 2014, Theor. Comput. Sci..

[18]  T. Kurtz.  Solutions of ordinary differential equations as limits of pure jump markov processes , 1970, Journal of Applied Probability.

[19]  Theodore L. Willke,et al.  A survey of inter-vehicle communication protocols and their applications , 2009, IEEE Communications Surveys & Tutorials.

[20]  Joost-Pieter Katoen,et al.  Quantitative Model Checking of Continuous-Time Markov Chains Against Timed Automata Specifications , 2009, 2009 24th Annual IEEE Symposium on Logic In Computer Science.

[21]  Enrico Bibbona,et al.  Strong approximation of density dependent Markov chains on bounded domains , 2017, 1704.07481.

[22]  Serge Haddad,et al.  Model Checking Timed and Stochastic Properties with CSL^{TA} , 2009, IEEE Transactions on Software Engineering.

[23]  Evsey Morozov.  Weak Regeneration in Modeling of Queueing Processes , 2004, Queueing Syst. Theory Appl..

[24]  Luca Bortolussi,et al.  Model checking single agent behaviours by fluid approximation , 2015, Inf. Comput..

[25]  Marco Beccuti,et al.  Analysis of Petri Net Models through Stochastic Differential Equations , 2014, Petri Nets.

[26]  Ronald W. Wolff,et al.  A Review of Regenerative Processes , 1993, SIAM Rev..

[27]  Evsey Morozov,et al.  A multiserver retrial queue: regenerative stability analysis , 2007, Queueing Syst. Theory Appl..

[28]  Luca Bortolussi,et al.  Model Checking Markov Population Models by Central Limit Approximation , 2013, QEST.

[29]  Paolo Ballarini,et al.  COSMOS: A Statistical Model Checker for the Hybrid Automata Stochastic Logic , 2011, 2011 Eighth International Conference on Quantitative Evaluation of SysTems.

[30]  Paolo Ballarini,et al.  HASL: A new approach for performance evaluation and model checking from concepts to experimentation , 2015, Perform. Evaluation.