PocketData
/
grant-NSF-CRI-2015
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

NSF bib style.

master
Geoffrey Challen 2015-11-06 10:27:20 -05:00
parent 9c9f7e9659
commit 3b3374db64
2 changed files with 1180 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1134
nsf.bst Normal file
View File

File diff suppressed because it is too large Load Diff

56
shortproposal.tex
View File

@ -138,27 +138,63 @@ differs from current benchmarks in four ways:
\section{Research Focus}
The limited power, cpu, and memory resources of smartphones provides strong motivation for research that makes small, personal-scale computing more efficient. In this proposal, we focus on enabling research on embedded databases specifically. Embedded databases bring a new array of challenges to the database field, and simultaneously represent a considerable part of a smartphone's resource consumption; An Android smartphone answers more than two queries per second using SQLite on average~\cite{pocketdata}.
The limited power, cpu, and memory resources of smartphones provides strong
motivation for research that makes small, personal-scale computing more
efficient. In this proposal, we focus on enabling research on embedded
databases specifically. Embedded databases bring a new array of challenges
to the database field, and simultaneously represent a considerable part of a
smartphone's resource consumption; An Android smartphone answers more than
two queries per second using SQLite on average~\cite{pocketdata}.
The first step in research on embedded databases is identifying efficiency bottlenecks, common usage patterns, and other specific opportunities for improvement. As host to the \PhoneLab project, UB is uniquely positioned to enable analysis of smartphone embedded database usage in the wild. We have already released a preliminary dataset and accompanying analysis~\cite{pocketdata}. If funded, we will be able to continue these efforts by releasing more, higher-quality data to the research community. Using this data, the research community will be able to identify insights into how embedded databases are used, and how they can be improved.
The first step in research on embedded databases is identifying efficiency
bottlenecks, common usage patterns, and other specific opportunities for
improvement. As host to the \PhoneLab project, UB is uniquely positioned to
enable analysis of smartphone embedded database usage in the wild. We have
already released a preliminary dataset and accompanying
analysis~\cite{pocketdata}. If funded, we will be able to continue these
efforts by releasing more, higher-quality data to the research community.
Using this data, the research community will be able to identify insights
into how embedded databases are used, and how they can be improved.
As opportunities for improvement are identified, it will become necessary to evaluate proposed solutions. Our second goal is to develop and maintain a benchmarking toolkit for pocket-scale data management. By providing our toolkit as a set of modules, we will be able to capture newly discovered limitations of existing pocket data management solutions in the benchmark. In this way, we hope to motivate further research on embedded data management solutions, and also to motivate embedded database developers to integrate these solutions into their products.
As opportunities for improvement are identified, it will become necessary to
evaluate proposed solutions. Our second goal is to develop and maintain a
benchmarking toolkit for pocket-scale data management. By providing our
toolkit as a set of modules, we will be able to capture newly discovered
limitations of existing pocket data management solutions in the benchmark.
In this way, we hope to motivate further research on embedded data management
solutions, and also to motivate embedded database developers to integrate
these solutions into their products.
\section{Sample Research Project}
In our preliminary analysis~\cite{pocketdata}, we noted that known limitations of object-relational mappers were causing a significant increase in the number of queries processed by SQLite. Compiler-based database integrations like LINQ~\cite{Meijer:2006:LRO:1142473.1142552}, LMS~\cite{Rompf:2015:FPS:2784731.2784760}, or StatusQuo~\cite{StatusQuo} begin to address these limitations. To tune and evaluate these solutions, it is necessary to have the real-world app query semantics that the PocketData project can provide.
In our preliminary analysis~\cite{pocketdata}, we noted that known
limitations of object-relational mappers were causing a significant increase
in the number of queries processed by SQLite. Compiler-based database
integrations like LINQ~\cite{Meijer:2006:LRO:1142473.1142552},
LMS~\cite{Rompf:2015:FPS:2784731.2784760}, or StatusQuo~\cite{StatusQuo}
begin to address these limitations. To tune and evaluate these solutions, it
is necessary to have the real-world app query semantics that the PocketData
project can provide.
\section{Community Involvement}
Our preliminary analysis~\cite{pocketdata} was presented to the Transactional Processing Council as part of an argument for an embedded database benchmark during their annual workshop.
The TPC is an organization that oversees many of the popular database benchmarks, and can provide feedback, publicity, and a deployment vector for our efforts.
Our argument was received favorably, and since then we have been working with the TPC leadership to obtain buy-in from smartphone vendors.
We have also been discussing our proposal with potential stakeholders. For example, Arnab Nandi from Ohio State has stated an interest in contributing trace data from his efforts on interactive tools for smartphone data management. We have also received further feedback in the form of desiderata from other faculty working on embedded databases. If funded, we will continue both of these outreach efforts.
Our preliminary analysis~\cite{pocketdata} was presented to the Transactional
Processing Council as part of an argument for an embedded database benchmark
during their annual workshop. The TPC is an organization that oversees many
of the popular database benchmarks, and can provide feedback, publicity, and
a deployment vector for our efforts. Our argument was received favorably,
and since then we have been working with the TPC leadership to obtain buy-in
from smartphone vendors.
We have also been discussing our proposal with potential stakeholders. For
example, Arnab Nandi from Ohio State has stated an interest in contributing
trace data from his efforts on interactive tools for smartphone data
management. We have also received further feedback in the form of desiderata
from other faculty working on embedded databases. If funded, we will
continue both of these outreach efforts.
{\footnotesize
\bibliographystyle{abbrv}
\bibliographystyle{nsf}
\bibliography{main}
}