Toward a verified relational database management system

Abstract

We report on our experience implementing a lightweight, fully verified relational database management system (RDBMS). The functional specification of RDBMS behavior, RDBMS implementation, and proof that the implementation meets the specification are all written and verified in Coq. Our contributions include: (1) a complete specification of the relational algebra in Coq; (2) an efficient realization of that model (B+ trees) implemented with the Ynot extension to Coq; and (3) a set of simple query optimizations proven to respect both semantics and run-time cost. In addition to describing the design and implementation of these artifacts, we highlight the challenges we encountered formalizing them, including the choice of representation for finite relations of typed tuples and the challenges of reasoning about data structures with complex sharing. Our experience shows that though many challenges remain, building fully-verified systems software in Coq is within reach.

Dynamic Boundaries: Information Hiding by Second Order Framing with First Order Assertions

“…We show by examples that our SOF rule handles not only invariants that pertain to several objects with a single owner but also design patterns in which several client-reachable peers cooperate and in which data structures may be overlapping or irregular. These are incompatible with ownership and remain as challenge problems in the current literature [4,22,27]. A program may link together multiple modules, each with its own hidden invariant and dynamic boundary.…”

“…Dropping the rule facilitates the modeling of higher order framing rules that capture something like visible state semantics for invariants even in programs using code pointers (e.g., [36]). The metatheory underlying the Ynot tool for interactive verification [27] uses a model that does not validate the conjunction rule [33]. Higher order separation logics offer elegant means to achieve data abstraction and strong functional specifications of interesting design patterns [20,19,27].…”

“…The metatheory underlying the Ynot tool for interactive verification [27] uses a model that does not validate the conjunction rule [33]. Higher order separation logics offer elegant means to achieve data abstraction and strong functional specifications of interesting design patterns [20,19,27]. The ability to explicitly quantify over invariants would seem to lessen the importance of hiding, but it requires considerable sophistication on the part of the user and her reasoning tools.…”

Toward an Information Infrastructure for Global Health Improvement

“…These new modes of thinking will be multi-disciplinary [68] and will invoke important concepts such as co-production [69,70]; "chaordic" organizations [71]; ultralarge-scale systems [72,73]; co-opetition [74,75]; evolution at whole-industry scale; social sensing [76]; learning over highly distributed, locally autonomous nodes [77]; and methods for developing ultra-high-assurance software-intensive systems, such as those employing emerging techniques of proof engineering [78,79,80].…”

A Coq Formalization of the Relational Data Model

“…This work investigates different formalizations of the unnamed version of the model and only addresses data definition and relational algebra aspects. A more recent formalization is found in Malecha et al, [8] which addresses the problem of designing a fully verified, lightweight implementation of a relational database system. The authors prove that their implementation meets the specification, all the proofs being written and verified in the Ynot [3] extension of Coq.…”