Search

Developers face a fundamental tension between performance and ease of programming when building complex applications. On the one hand, by freeing developers from having to worry about concurrency and failures, ACID transactions can greatly simplify the task of building applications. These benefits, however, have a cost: in a distributed setting, ACID transactions can limit performance and make systems hard to scale in the presence of contention. On the other hand, the BASE approach can achieve higher performance by providing weakened semantics. Embracing the BASE paradigm, however, exacts its own heavy price: once one renounces consistency guarantees, it is up to developers to explicitly code in their applications the logic necessary to en- sure consistency in the presence of concurrency and faults, making this task complex and error-prone. This dissertation aims to resolve this tension by building database systems that provide both the ease of programming of the ACID approach and the performance that the BASE approach can provide. Our approach depends on the observation that different transactions affect overall performance of applications in different ways. Traditional ACID databases ignore this diversity: they provide a single implementation of the same uniform abstraction across all transactions. This dissertation explores two different ways to leverage the previews key observation to combine performance and ease of programming, and presents two systems, Salt and Callas, inspired by them.

Two-dimensional free surface flows may be formulated through boundary integrals which allows analytic continuation in the unphysical complex plane. For two dimensional water waves, Tanveer showed that the only form of singularity in the unphysical plane is of the square-root type. One may wonder what influence these singularities may have on the behavior of water waves. This thesis resolves some of these questions by tracking singularities in the unphysical domain and relating their close approach to the real axis with wave breaking.The main result is the direct verification of Tanveer’s singularity result. A boundaryintegral technique is used to simulate deep water wave motion. A spectral procedure is used to form-fit the Fourier spectrum of the curvature of the wave profile to a prescribed asymptotic expression. The form-fit provides information on the power and location of the closest singularity to the real axis. The power of the curvature singularities is found to be -3/2 when the curvature is expressed as a function of the Lagrangian variable. This singularity is associated with a pole singularity in the complex arclength plane, and is not an artifact of the parametrization. The singularity approaches the real axis when a plunging breaker occurs. For nonbreaking waves, the singularity wanders above some level in the unphysical plane. It is then established that this curvature singularity is theoretically equivalent to Tanveer’s one-half power singularity. When the surface elevation is viewed as a function of horizontal distance, a different type of singularity arises. It is a square root type singularity that takes the form of a breaking wave when it reaches the real axis of the horizontal coordinate.Nonlinear interactions among various wavelengths are considered important in random ocean waves. A particularly important nonlinear interaction is the Benjamin-Feir instability. For moderate initial amplitudes, the end-state of this instability is either wave breaking or the Fermi-Pasta-Ulam recurrence. Clearly, singularities in the unphysical domain will play a role. Starting with initial conditions that contain several singularities away from the real axis, their trajectories are studied. When breaking occurs, results show that it is one of the crests in the wave train that breaks like a plunging breaker while others remain moderately flat. One of the singularities moves close to the real axis while the other singularities stay far away. When recurrence occurs, evidence indicates that the closest singularity remains far away from the real axis for all time. The hope and the possibility is that a study of singularities in Benjamin-Feir instability may lead to insight into nonlinear wave interactions in general.