11.01.2017 Views

A Technical History of the SEI

ihQTwP

ihQTwP

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

The benefits <strong>of</strong> <strong>the</strong> IMPACT methodology and associated tools to DoD systems were made apparent<br />

in two demonstrations: a phased array radar scheduling demonstration and an embedded avionics<br />

simulation designed to support <strong>the</strong> F-22 and F-35 JSF programs.<br />

The first demonstration (at <strong>the</strong> DASADA Demonstration and Exposition in July 2002) showed <strong>the</strong><br />

runtime use <strong>of</strong> Q-RAM. A visual demonstration represented a carrier under attack from a set <strong>of</strong><br />

enemy locations. With a variety <strong>of</strong> threats attacking <strong>the</strong> ship, <strong>the</strong> phased-array radar had to be<br />

scheduled in such a way that <strong>the</strong> ship could destroy those threats. The radar, treated as a nonpreemptible<br />

resource, could be allocated to tracking targets in varying amounts to achieve different<br />

levels <strong>of</strong> service quality or tracking error [Hansen 2004]. The mission-critical embedded system<br />

demonstration illustrated <strong>the</strong> use <strong>of</strong> RTQT to assess <strong>the</strong> ability <strong>of</strong> a system to meet <strong>the</strong> timing<br />

requirements <strong>of</strong> stochastic task sets (such as mission-awareness applications) under heavy workload<br />

conditions. When RTQT indicates that latency requirements are not met, it provides an engineering<br />

basis for modifying <strong>the</strong> workload to achieve <strong>the</strong> desired latency within a specified degree<br />

<strong>of</strong> certainty.<br />

The Consequence: Bringing an Analytic Basis to Engineering Dynamic<br />

Systems<br />

Work on <strong>the</strong> F-22 embedded avionics simulation provided new and valuable temporal and performance<br />

analysis information to F-22 designers that was not previously available and that could be<br />

used in <strong>the</strong> final refinement <strong>of</strong> <strong>the</strong> F-22 weapon system. Fur<strong>the</strong>r, it provided insight into several<br />

design improvements for legacy systems. System utilization, resource efficiency, and maintainability<br />

could be significantly enhanced using DASADA-developed design practices. Progress made<br />

under DASADA allowed LM-Aero and CMU to propose a large-scale DASADA II experiment<br />

centered on upgrading <strong>the</strong> F-22 mission computer temporal architecture. Results and insights from<br />

this experiment aimed to reduce both new development and application rehost costs through <strong>the</strong><br />

analytic composition methodology developed under DASADA I. Additional benefits included enhanced<br />

adaptability to dynamic mission demands, along with increased system reconfiguration<br />

options.<br />

In addition, IMPACT technology was applied in <strong>the</strong> context <strong>of</strong> a rotorcraft through collaborations<br />

with U.S. Army Aviation and Missile Command (AMCOM). Investigations into integration <strong>of</strong> Q-<br />

RAM with hard real-time scheduling in avionics applications under DASADA led to interest in<br />

this technology by <strong>the</strong> U.S. Army AMCOM. This led to <strong>the</strong> invitation to join a team to propose a<br />

large-scale DASADA II experiment centered on application <strong>of</strong> DASADA technologies to <strong>the</strong> Sikorsky<br />

Black Hawk helicopter.<br />

The <strong>SEI</strong> Contribution<br />

This research was conducted in collaboration with <strong>the</strong> Carnegie Mellon faculty and Lockheed<br />

Martin Aeronautics Co. All members <strong>of</strong> <strong>the</strong> team made contributions to <strong>the</strong> IMPACT project.<br />

Since all results are those <strong>of</strong> <strong>the</strong> full team, it is not possible to isolate <strong>the</strong> specific contributions<br />

made by <strong>the</strong> <strong>SEI</strong>.<br />

CMU/<strong>SEI</strong>-2016-SR-027 | SOFTWARE ENGINEERING INSTITUTE | CARNEGIE MELLON UNIVERSITY 52<br />

Distribution Statement A: Approved for Public Release; Distribution is Unlimited

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!