Sandia’s approach to achieve and sustain operational excellence includes the routine application of Plan-Do-Check-Act (P-D-C-A) quality principles and defect prevention methodologies to our work. To sustain operational excellence, we need to be mindful of factors that could prevent success, address them before they do, and continue to improve all facets of our work. The mindful approach used by organizations that have achieved operational excellence includes use of a questioning attitude coupled with critical thinking throughout their workflows, and the application of five key principles: (1) a preoccupation with failure, (2) reluctance to oversimplify, (3) sensitivity to operations, (4) commitment to resilience, and (5) appropriate deference to expertise.
The microelectronics industry seeks screening tools that can be used to verify the origin of and track integrated circuits (ICs) throughout their lifecycle. Embedded circuits that measure process variation of an IC are well known. This paper adds to previous work using these circuits for studying manufacturer characteristics on final product ICs, particularly for the purpose of developing and verifying a signature for a microelectronics manufacturing facility (fab). We present the design, measurements and analysis of 159 silicon ICs which were built as a proof of concept for this purpose. 80 copies of our proof of concept IC were built at one fab, and 80 more copies were built across two lots at a second fab. Using these ICs, our prototype circuits allowed us to distinguish these two fabs with up to 98.7% accuracy and also distinguish the two lots from the second fab with up to 98.8% accuracy.
Self-heating induced failure of graphene devices synthesized from both chemical vapor deposition (CVD) and epitaxial means is compared using a combination of infrared thermography and Raman imaging. Despite a larger thermal resistance, CVD devices dissipate >3x the amount of power before failure than their epitaxial counterparts. The discrepancy arises due to morphological irregularities implicit to the graphene synthesis method that induce localized heating. Morphology, rather than thermal resistance, therefore dictates power handling limits in graphene devices.
As supercomputers move to exascale, the number of cores per node continues to increase, but the I/O bandwidth between nodes is increasing more slowly. This leads to computational power outstripping I/O bandwidth. This growth, in turn, encourages moving as much of an HPC workflow as possible onto the node in order to minimize data movement. One particular method of application composition, enclaves, co-locates different operating systems and runtimes on the same node where they communicate by in situ communication mechanisms. In this work, we describe a mechanism for communicating between composed applications. We implement a mechanism using Copy onWrite cooperating with XEMEM shared memory to provide consistent, implicitly unsynchronized communication across enclaves. We then evaluate this mechanism using a composed application and analytics between the Kitten Lightweight Kernel and Linux on top of the Hobbes Operating System and Runtime. These results show a 3% overhead compared to an application running in isolation, demonstrating the viability of this approach.
As supercomputers move to exascale, the number of cores per node continues to increase, but the I/O bandwidth between nodes is increasing more slowly. This leads to computational power outstripping I/O bandwidth. This growth, in turn, encourages moving as much of an HPC workflow as possible onto the node in order to minimize data movement. One particular method of application composition, enclaves, co-locates different operating systems and runtimes on the same node where they communicate by in situ communication mechanisms. In this work, we describe a mechanism for communicating between composed applications. We implement a mechanism using Copy onWrite cooperating with XEMEM shared memory to provide consistent, implicitly unsynchronized communication across enclaves. We then evaluate this mechanism using a composed application and analytics between the Kitten Lightweight Kernel and Linux on top of the Hobbes Operating System and Runtime. These results show a 3% overhead compared to an application running in isolation, demonstrating the viability of this approach.
