﻿<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/">
	<channel>
		
		<title>News - Institut für Technische Informatik</title>
		<link>http://www.iti.uni-stuttgart.de/</link>
		<description>News - Institut für Technische Informatik</description>
		<language>de</language>
		<image>
			<title>News - Institut für Technische Informatik</title>
			<url>http://www.iti.uni-stuttgart.de/fileadmin/templates/ITI/rss/ITI_194x116.png</url>
			<link>http://www.iti.uni-stuttgart.de/</link>
			<width>194</width>
			<height>116</height>
			<description>News - Institut für Technische Informatik</description>
		</image>
		<generator>TYPO3 - get.content.right</generator>
		<docs>http://blogs.law.harvard.edu/tech/rss</docs>
		
		
		
		<lastBuildDate>Thu, 19 Jul 2012 15:45:00 +0200</lastBuildDate>
		
		
		<item>
			<title>Efficient Simulation-based Characterization of Single Event Transients in Floating-Point Data Paths</title>
			<link>http://www.iti.uni-stuttgart.de/aktuell/details/efficient-simulation-based-characterization-of-single-event-transients-in-floating-point-data-paths/3494da7f60fcea613ee51d6e1a72a4ed.html</link>
			<description>15:45-16:30, V47.06,
Dipl.-Inform. Claus Braun, 
Institut für Technische Informatik </description>
			<content:encoded><![CDATA[Modern semiconductor devices like multicore CPUs or GPGPUs are  increasingly prone to transient errors in the logic. If the effects of  such transient events propagate into memory elements, internal states  within data paths , as well as user and program data can be compromised.  For the selection and dimensioning of suitable fault-tolerance  countermeasures, such effects have to be characterized across all  levels, from the hardware up to the software and the algorithm.<br /><br />Floating-point  arithmetic is an essential component of most applications in scientific  computing and simulation technology. For GPGPU accelerators,  floating-point processing is even the main operational task. In this  talk, a standard-compliant Floating-point Processing Unit (FPU) is used  to investigate the impact of soft errors in the data path. How  vulnerable is such a data path? What kinds of errors become visible and  how critical are they?]]></content:encoded>
			<category>Open Seminar - Rechnerarchitektur</category>
			<category>SS 12</category>
			
			
			<pubDate>Thu, 19 Jul 2012 15:45:00 +0200</pubDate>
			
		</item>
		
		<item>
			<title>Modeling of Design-for-test infrastructure in complex Systems-on-chips</title>
			<link>http://www.iti.uni-stuttgart.de/aktuell/details/modeling-of-design-for-test-infrastructure-in-complex-systems-on-chips/965cef6636395fcdec2aa581cd0bda97.html</link>
			<description>15:45-16:30, V47.06,
M. Sc. cand. David Buntoro, 
Institut für Technische Informatik </description>
			<content:encoded><![CDATA[Every integrated circuit contains a piece of design-for-test (DFT)  infrastructure in order to guarantee the chip quality after manufacture.  The DFT resources are employed only once in the fab and are usually not  available during regular system operation.<br /><br />In order to assess  the hardware integrity of a chip over its complete life-cycle, it is  promising to reuse the DFT infrastructure as part of system-level test.<br /><br />In  this thesis, the provided system, a Tricore processor from Infineon,&nbsp;  must me partitioned and modified in order to enable the autonomous  structural&nbsp; test of every component of the system in the field without  expensive external equipment.]]></content:encoded>
			<category>Open Seminar - Rechnerarchitektur</category>
			<category>SS 12</category>
			
			
			<pubDate>Thu, 12 Jul 2012 15:45:00 +0200</pubDate>
			
		</item>
		
		<item>
			<title>Effectivity of high-level mutants in lower level fault representation</title>
			<link>http://www.iti.uni-stuttgart.de/aktuell/details/effectivity-of-high-level-mutants-in-lower-level-fault-representation/d6968302e875b09a7b6ae602520603c5.html</link>
			<description>16:30-17:15, V47.06, 
Dipl.-Inf. Laura Rodríguez Gómez,
Institut für Technische Informatik</description>
			<content:encoded><![CDATA[Mutation testing is gaining importance as a method for system level test generation. Although a wide range of different mutants has been developed for system level, it is still unclear that test data generated to achieve a high mutation score also performs well when looking for high fault coverage. Thus, the correlation between high level mutants and lower level fault models must be studied to determine if it is possible to reuse high level test data for lower level fault detection.]]></content:encoded>
			<category>Open Seminar - Rechnerarchitektur</category>
			<category>SS 12</category>
			
			
			<pubDate>Thu, 05 Jul 2012 16:30:00 +0200</pubDate>
			
		</item>
		
		<item>
			<title>Aging Monitor: Path Selective Monitoring</title>
			<link>http://www.iti.uni-stuttgart.de/aktuell/details/aging-monitor-path-selective-monitoring/b6ab394ebd0061c085c252d9f09e6c0c.html</link>
			<description>15:45-16:30, V47.06,
M. Sc. Chang Liu,
Institut für Technische Informatik</description>
			<content:encoded><![CDATA[The transistor dimension scaling down brings us new challenging for reliability issues. As one of the major concerns, transistors’ aging necessitates implementing online aging monitors to observe the chips performance. Today’s aging monitors include Active and Dormant sensors. Due to the lack of work load information, all active aging sensors are pessimistic. Dormant monitors overcome this problem however the comparative large hardware overhead makes it impossible to apply the aging sensor to every flip-flop in the target circuit. Thus there is a chance that the longest paths with aging checkers aren’t sensitized very often and the aging progress is not monitored. <br />In this seminar, to avoid above unwanted situation, two solutions are proposed: stability checker relocation and path activation monitor development. The efficiency of these synergic solutions is proved by the experimental results.]]></content:encoded>
			<category>Open Seminar - Rechnerarchitektur</category>
			<category>SS 12</category>
			
			
			<pubDate>Thu, 05 Jul 2012 15:45:00 +0200</pubDate>
			
		</item>
		
		<item>
			<title>Tag der Wissenschaften</title>
			<link>http://www.iti.uni-stuttgart.de/aktuell/details/tag-der-wissenschaften/3adf6f7503d93b6dfbaf0002fa5c94c0.html</link>
			<description>11:00 - 19:00 Uhr, Campus Vaihingen</description>
			<content:encoded><![CDATA[Alle weiteren Informationen erhalten Sie unter diesem <link http://www.uni-stuttgart.de/tag/2012/ _blank>Link</link>.]]></content:encoded>
			<category>Events</category>
			
			
			<pubDate>Sat, 30 Jun 2012 11:00:00 +0200</pubDate>
			
		</item>
		
		<item>
			<title>Emulation of automotive complex systems-on-chips</title>
			<link>http://www.iti.uni-stuttgart.de/aktuell/details/emulation-of-automotive-complex-systems-on-chips/e6eec1b0586ff29162882275708a6a02.html</link>
			<description>16:30-17:15, V47.06,
M. Sc. cand. Mina Tawfik, 
Institut für Technische Informatik </description>
			<content:encoded><![CDATA[Emulation of automotive complex systems-on-chipsThe rising  complexity of today’s systems such as automotive controllers requires  the thorough validation of its software and hardware components. As the  computational effort for simulating such large systems takes an  increasing amount of time, emulation techniques have been gaining  importance. Rapid prototyping with FPGAs provides an affordable and fast  solution for the validation of large software/hardware systems. <br /><br />In  this thesis, an emulation system for a target automotive controller  must be architected and implemented. The controller is the Tricore, a  current microprocessor design provided by our industrial partners.]]></content:encoded>
			<category>Open Seminar - Rechnerarchitektur</category>
			<category>SS 12</category>
			
			
			<pubDate>Thu, 28 Jun 2012 16:30:00 +0200</pubDate>
			
		</item>
		
		<item>
			<title>Test pattern Generation for S/H time violations under variations</title>
			<link>http://www.iti.uni-stuttgart.de/aktuell/details/test-pattern-generation-for-sh-time-violations-under-variations/82160003e8277772a5e8156203a2f9d0.html</link>
			<description>15:45-16:30, V47.06, 
M.Sc., Anusha Kakarala, 
Institut für Technische Informatik</description>
			<content:encoded><![CDATA[]]></content:encoded>
			<category>Open Seminar - Rechnerarchitektur</category>
			<category>SS 12</category>
			
			
			<pubDate>Thu, 28 Jun 2012 15:45:00 +0200</pubDate>
			
		</item>
		
		<item>
			<title>Efficient Multi-Level Parallel NFP Simulation Platform </title>
			<link>http://www.iti.uni-stuttgart.de/aktuell/details/efficient-multi-level-parallel-nfp-simulation-platform/ceb133d741ccac10d718e2981f06ab5b.html</link>
			<description>16:30-17:15, V47.06, 
M.Sc. Nadereh Hatami, 
Institut für Technische Informatik</description>
			<content:encoded><![CDATA[Non-functional properties (NFPs) of integrated circuits include  reliability, vulnerability, power consumption or heat dissipation.  Accurate NFP prediction over long periods of system operation poses a  great challenge due to prohibitive simulation costs. For instance, in  case of aging estimation, the existing low-level models are accurate but  not efficient enough for simulation of complex designs. On the other  hand, existing <br />techniques for fast high-level simulation do not provide enough details  for NFP analysis. The goal of this paper is to bridge this gap by  combining the accuracy of low-level models with high-level simulation  speed. We introduce an efficient mixed-level NFP prediction methodology  that considers both the structure and application of a system. The  system is modeled at transaction-level to enable high simulation speed.  To maintain accuracy, NFP assessment for cores under analysis is  conducted at gate-level by cycle-accurate simulation. To speed up the  simulation, high level transactions are being simulated in parallel  using gate-level state prediction. We propose effective techniques for  cross-level synchronization and idle simulation speed-up. The proposed  method is from 7 up to 400 times faster with mean error below 0.006%. ]]></content:encoded>
			<category>Open Seminar - Rechnerarchitektur</category>
			<category>SS 12</category>
			
			
			<pubDate>Thu, 21 Jun 2012 16:30:00 +0200</pubDate>
			
		</item>
		
		<item>
			<title>Optimal Pattern Retargeting for Reconfigurable Scan Networks</title>
			<link>http://www.iti.uni-stuttgart.de/aktuell/details/optimal-pattern-retargeting-for-reconfigurable-scan-networks/249b7e4370fe0df741f53f7002b52917.html</link>
			<description>15:45-16:30, V47.04,
M.Sc. Rafal Baranowski, 
Institut für Technische Informatik</description>
			<content:encoded><![CDATA[]]></content:encoded>
			<category>Open Seminar - Rechnerarchitektur</category>
			<category>SS 12</category>
			
			
			<pubDate>Thu, 21 Jun 2012 15:45:00 +0200</pubDate>
			
		</item>
		
		<item>
			<title>In-Field Scan-Test of Automotive ICs</title>
			<link>http://www.iti.uni-stuttgart.de/aktuell/details/in-field-scan-test-of-automotive-ics/3c101cf30ea9ad8ac96c4750b8049f7c.html</link>
			<description>16:15-16:45, V47.06, 
Dipl.-Inf. Dominik Ull, 
Institut für Technische Informatik</description>
			<content:encoded><![CDATA[]]></content:encoded>
			<category>Open Seminar - Rechnerarchitektur</category>
			<category>SS 12</category>
			
			
			<pubDate>Thu, 24 May 2012 16:15:00 +0200</pubDate>
			
		</item>
		
	</channel>
</rss>
