A robust variant of block Jacobi-Davidson for extracting a large number of eigenpairs: Application to grid-based real-space density functional theory.

In this work, we investigate a block Jacobi-Davidson (J-D) variant suitable for sparse symmetric eigenproblems where a substantial number of extremal eigenvalues are desired (e.g., ground-state real-space quantum chemistry). Most J-D algorithm variations tend to slow down as the number of desired eigenpairs increases due to frequent orthogonalization against a growing list of solved eigenvectors. In our specification of block J-D, all of the steps of the algorithm are performed in clusters, including the linear solves, which allows us to greatly reduce computational effort with blocked matrix-vector multiplies. In addition, we move orthogonalization against locked eigenvectors and working eigenvectors outside of the inner loop but retain the single Ritz vector projection corresponding to the index of the correction vector. Furthermore, we minimize the computational effort by constraining the working subspace to the current vectors being updated and the latest set of corresponding correction vectors. Finally, we incorporate accuracy thresholds based on the precision required by the Fermi-Dirac distribution. The net result is a significant reduction in the computational effort against most previous block J-D implementations, especially as the number of wanted eigenpairs grows. We compare our approach with another robust implementation of block J-D (JDQMR) and the state-of-the-art Chebyshev filter subspace (CheFSI) method for various real-space density functional theory systems. Versus CheFSI, for first-row elements, our method yields competitive timings for valence-only systems and 4-6× speedups for all-electron systems with up to 10× reduced matrix-vector multiplies. For all-electron calculations on larger elements (e.g., gold) where the wanted spectrum is quite narrow compared to the full spectrum, we observe 60× speedup with 200× fewer matrix-vector multiples vs. CheFSI.

Advancing a distributed multi-scale computing framework for large-scale high-throughput discovery in materials science.

We describe the development of a large-scale high-throughput application for discovery in materials science. Our point of departure is a computational framework for distributed multi-scale computation. We augment the original framework with a specialized module whose role is to route evaluation requests needed by the high-throughput application to a collection of available computational resources. We evaluate the feasibility and performance of the resulting high-throughput computational framework by carrying out a high-throughput study of battery solvents. Our results indicate that distributed multi-scale computing, by virtue of its adaptive nature, is particularly well-suited for building high-throughput applications.

Standard and alternative methods of stretcher carriage: performance, human factors, and cardiorespiratory responses.

Transporting a casualty on a stretcher is a common task for medical and military personnel. Stretchers are usually carried by hand, but distributing the load to other parts of the body may have advantages. To examine alternative carriage methods, 11 soldiers walked on a treadmill at 4.8 km/h while performing two-person carries of a stretcher containing an 80-kg manikin. In separate trials, soldiers carried the stretcher using: (1) hand carriage, (2) shoulder straps, (3) a specially designed harness that allowed load shifting between the hips and shoulders (hip-shoulder system), and (4) a clip that fitted on the belt of standard military load carrying equipment (LCE) and placed the stretcher mass mainly on the hips. With each system, subjects walked until volitional fatigue or 30 min. While walking, expired gases and heart rates were obtained and subjects rated their perceived exertion (Borg Scale). At the conclusion of all four trials, subjects rated each system on a number of subjective measures. Results showed that average (+/- SD) carriage times were 2.7+/-1.4, 14.5+/-8.3, 25.4+/-8.1, and 21.7+/-9.9 min with the hand, shoulder, hip-shoulder and LCE systems respectively (p<0.01). Hand carriage resulted in considerably more cardiorespiratory stress (higher heart rate and minute ventilation, p<0.05) than the other three systems, but there were few consistent differences among the other three systems. Perceived exertion in the upper body was less with the hip-shoulder and LCE systems than with the other two systems (p<0.05). Subjects preferred the hip-shoulder and LCE systems overall and for specific subjective characteristics such as comfort, ease of use and stability (p<0.01). These data indicate that moving the stretcher load from the hands and placing that load on the shoulders and/or hips results in improved performance, reduced cardiorespiratory stress and favourable subjective evaluations. Further developmental work should focus on the hip-shoulder and LCE systems.

Welfare reform: advocacy and intervention in the health care setting.

Welfare reform has drastically altered the lives of poor families in the US. In its wake, many former recipients are not receiving whatever transitional benefits and other safeguards to which they remain entitled under federal and state laws. Families are losing access to Medicaid and are not receiving the child care assistance or Food Stamps for which they continue to be eligible. Ill-served by stringent time limits and work requirements, lack of child care assistance, and lack of training and educational opportunities for the development of skills that will lead to better jobs, families need help to navigate the complexities of the new welfare system. Boston Medical Center's Department of Pediatrics has instituted a welfare screening project to educate families about their rights under welfare reform and assist them in advocating for themselves and their children.