Page 26 - FIGI: Security Aspects of Distributed Ledger Technologies
P. 26

Figure 4: DLT architecture abstraction layers 134





























            A: Network layer: Decentralized communication model
            B: Data model layer - The structure, content, and the operation of the DLT data.
            C: Consensus layer - Where all nodes in the DL attempt to agree on the content to be added to the DLT
            D: Execution layer - Contains details of the runtime environment that support DLT operations. Each DLT system uses its
            own type.
            E: Application layer - Includes the use-cases of the DLT application.
            F: External layer- All the external input/outputs into a DLT and/or use of tokens on a DLT
            chain ecosystem – these IoTs acting as DLT oracles   Annex D summarizes these general risks and vul-
            are often not secure and create the opportunity for   nerability concerns, alongside resultant risks and
            injection of incorrect data in a DLT that could set off a   potential  mitigation measures.  Other  areas  of  con-
            chain of incorrect smart contract ‘transactions.’ Zero-  cern are described in Table 5 and include ‘download
            knowledge-proof can solve this issue, since the nodes   and decrypt later’ concerns; (un)authorized access;
            can validate the authenticity of the data injected by   increased nodes increase vulnerabilities; interopera-
            the oracles without gaining access to the data itself.  bility attempts between DLTs; open source software
               As noted above on methodology used in this      development in DLTs; trust of nodes; user interface/
            study, to illustrate the loci of the attacks from threat   user experience failures; and privacy and confidenti-
            vectors we use an adapted version of a published    ality of data.
                                                         133
            DLT architecture abstraction layers which are based
            on a layered DLT architecture approach. These      8�2  Software Development Flaws
            abstract layers consist of a network layer, a data lay-
            er, a consensus layer, an execution layer, and an appli-  8.2.1   Issue: Methods to speed up DLT
            cation layer, and an external layer. These layers are   transaction processing may be insecure
            shown in Figure 4�                                 Many public, permissionless blockchain aspire to
               These dimensions are integrated into the most   achieve a fully decentralized operation.  The block-
                                                                                                 135
            prominent threats and vulnerabilities that this report   chain scalability trilemma  represents a widely held
                                                                                     136
            identifies as having the most coincidence to finan-  belief that the use of blockchain technology presents
            cial inclusion. As shown in Figure 5, these prominent   a tri-directional compromise in efforts to increase
            risks and vulnerabilities include software develop-  scalability, security and decentralization.  All three
                                                                                                   137
            ment flaws; DLT availability; transaction and data   cannot be maximized at one time and increasing the
            accuracy;  key  management;  data  privacy  and  pro-  level of one factor results in the decrease of another.
            tection; safety of funds; consensus; smart contracts.   Hence blockchain’s goals of striving to reach maxi-
            Annex D combines these layers, risk, threats and vul-  mum levels of decentralization inherently result in a
            nerabilities.



           24    Security Aspects of Distributed Ledger Technologies
   21   22   23   24   25   26   27   28   29   30   31