This is the home page for our 23rd major research exhibit at the IEEE/ACM
Supercomputing conference. The exhibit is now under the slightly changed
title University of Kentucky / Aggregate.Org, but it
is still really run by our informal research consortium led by our KAOS
(Compilers, Hardware Architectures, and Operating Systems) group here at the
University of Kentucky's Department
of Electrical and Computer Engineering -- hopefully, for the last time.
The Center for Computational Sciences is displaying various things in the
exhibit this year, and is expected to take the lead for future exhibits. We
are booth #3830, a 10x20 booth; the exhibit hall essentially
has two wings, and we are located about 1/3 of the way up in the right wing
(Hall 2 in this map).
Our 23rd Exhibit
As usual, our research exhibit this year will be showing a lot
of different things we have developed or are developing...
ranging from supercomputer architecture to 3D printing technology
and a new model for imaging sensors.
We kept handouts to just four topics:
TIK: Temporal Imaging from Kentucky -- for
years we've been working on Time Domain Continuous Imaging (TDCI), which
replaces the concept of am imager capturing frames with capture of continuous
waveforms for pixel light level over time. With TIK, you can do this using
conventional still and video cameras!
Making Time Tooling Around --
Perhaps you're wondering what the computer engineering research group that
created PCCTS/Antlr, SWAR (SIMD Within A Register), the first Linux PC cluster
supercomputer, FNNs (Flat Neighborhood Networks), and MOG (MIMD On GPU) is
doing playing with machine tools? We began using these machines to make camera
parts for our computational photography research, but we're really about
making the hardware and systems software components of a computing system work
better together, and those machine tools are computer systems. Never forget
that you can use your tools to make better tools....
Performance-Engineered Computer Networks --
an overview of the many tools and technologies we have developed to improve
computer networks, especially those within parallel supercomputers. AFNs,
FNNs, KNITT, Netwires, and NodeScape are all briefly described. (Just last
year, we learned that Google has been using our FNN work as the basis for
designing much of their network infrastructure.)
A Maze of Twisty Little Passages --
an update to MOG (MIMD on GPU). The MOG core is becoming close to
production quality and there are now several system calls implemented
(the GPU code can now perform host system calls). We didn't bring
the big wooden maze, but did bring a couple of small plastic ones that
made the MIMD-can-run-efficiently-on-SIMD point as well as their
giant wooden sibling used to.
Here's what our booth looks like this year:
and here's a time-lapse of the entire show:
3D printing plays a major role in the computational photography
work we are doing (for 3D printing custom camera parts), and 3D
printing is cool, but we didn't want to risk bringing our MakerGear M2
as airline checked baggage. Thus, we bought and brought a $400
Wanhao I3. Really quite a nice machine... more finished than the M2,
and capable of comprable print quality, but at much slower