The rapidly growing number of large network analysis problems has led to the emergence of many parallel and distributed graph processing systems—one survey in 2014 identified over 80. Since then, the landscape has evolved; some packages have become inactive while more are being developed. Determining the best approach for a given problem is infeasible for most developers.
Directed Research Project
In both high-performance computing and Internet of Things (IoT) applications, it is important for programmers to utilize the hardware as efficiently as possible in terms of both performance and power usage. Automatic methods of tuning code for such efficiency have been developed and used for many years in HPC, but less so in other areas. In this paper, we apply these methods to the domain of IoT to explore the feasibility of using autotuning in this domain. Our case study examines the application of roadway traffic analysis.
Dynamic specification mining techniques attempt to fill gaps in missing or decaying documentation of software systems to support software maintenance tasks such as testing or bug fixing. Current dynamic mining techniques are blind to common coding styles, and in particular to design patterns that involve dynamic data structures such as lists of listeners for event notification.
Peer to Peer (P2P) systems offer a promising approach for distributed data management in a scalable fashion. Distributed Hash Table (DHT) is a subclass of P2P systems, which is often referred to as structured P2P system. DHTs have been an active area of research since 2001 but have not been deployed in a real world setting only until recently. As a result, characteristics of a widely deployed DHT are not well understood. Characterizing a DHT requires capturing the global view of the P2P system which is challenging due to the large size, and distributed and dynamic nature of P2P systems.
Transactional programming models have been proposed as an important part of the evolution of threaded programming in shared memory environments; however, a critical shortcoming of these models is their inability to function and compose properly in the presence of operations that do not have a transactional nature (e.g., system calls).
The ontology-based information integration is a popular topic in the Semantic Web and database community. Even in the same domain, different groups of domain experts define different ontologies to share data. Therefore data integration and query translation between ontologies become main topics. Since fully automatic ontology mapping is an AI-complete problem, my advisor, Prof. Dou, pointed out that we should find an approach to solve this problem with less human involvement to save human effort.
ZFIN and MGI are two research groups in the field of bioinformatics that collect and distribute genetic data regarding zebrafish (Danio rerio) and mouse (Mus musculus) respectively. They provide their own web interfaces to search their data, which are stored in relational database systems. This project was aimed at developing web interfaces that are capable of searching across both databases. This is useful to biologists to discover patterns that span the two species in areas such as gene expressions and phenotypes.
Advances in computing and communication through internet and Web have resulted in many pervasive distributed information resources which in general are structurally and semantically heterogeneous even in the same domain. However, heterogeneity itself has not been studied in a formal way so that the representation of different kinds of heterogeneities can be generically processed by other programs automatically. Most descriptions and categorization schemes of data heterogeneities are given in natural language.
Workflows offer scientists a simple but flexible programming model at a level of abstraction closer to the domain-specific activities that they seek to perform. However, languages for describing workflows tend to be highly complex, or specialized towards a particular domain, or both. WOOL is an abstract workflow language with human-readable syntax, intuitive semantics, and a powerful abstract type system. WOOL workflows can be targeted to almost any kind of runtime system supporting data-flow computation.
It has been shown that a language with the delimited control operators shift and reset can be translated into a language containing only an undelimited control operator such as callcc and a single mutable storage cell. My Directed Research Project with my advisor, Professor Zena Ariola, studied this encoding and a related one from a language containing callcc into one containing the control and prompt operators. We showed that the translation does not faithfully preserve other control effects that may be present. We developed improved encodings in response to these problems.