IBM

About: IBM is a company organization based out in Armonk, New York, United States. It is known for research contribution in the topics: Layer (electronics) & Signal. The organization has 134567 authors who have published 253905 publications receiving 7458795 citations. The organization is also known as: International Business Machines Corporation & Big Blue.

An Analysis of Efficient Multi-Core Global Power Management Policies: Maximizing Performance for a Given Power Budget

Abstract: Chip-level power and thermal implications will continue to rule as one of the primary design constraints and performance limiters. The gap between average and peak power actually widens with increased levels of core integration. As such, if per-core control of power levels (modes) is possible, a global power manager should be able to dynamically set the modes suitably. This would be done in tune with the workload characteristics, in order to always maintain a chip-level power that is below the specified budget. Furthermore, this should be possible without significant degradation of chip-level throughput performance. We analyze and validate this concept in detail in this paper. We assume a per-core DVFS (dynamic voltage and frequency scaling) knob to be available to such a conceptual global power manager. We evaluate several different policies for global multi-core power management. In this analysis, we consider various different objectives such as prioritization and optimized throughput. Overall, our results show that in the context of a workload comprised of SPEC benchmark threads, our best architected policies can come within 1% of the performance of an ideal oracle, while meeting a given chip-level power budget. Furthermore, we show that these global dynamic management policies perform significantly better than static management, even if static scheduling is given oracular knowledge.

To prune, or not to prune: exploring the efficacy of pruning for model compression

Abstract: Model pruning seeks to induce sparsity in a deep neural network's various connection matrices, thereby reducing the number of nonzero-valued parameters in the model. Recent reports (Han et al., 2015; Narang et al., 2017) prune deep networks at the cost of only a marginal loss in accuracy and achieve a sizable reduction in model size. This hints at the possibility that the baseline models in these experiments are perhaps severely over-parameterized at the outset and a viable alternative for model compression might be to simply reduce the number of hidden units while maintaining the model's dense connection structure, exposing a similar trade-off in model size and accuracy. We investigate these two distinct paths for model compression within the context of energy-efficient inference in resource-constrained environments and propose a new gradual pruning technique that is simple and straightforward to apply across a variety of models/datasets with minimal tuning and can be seamlessly incorporated within the training process. We compare the accuracy of large, but pruned models (large-sparse) and their smaller, but dense (small-dense) counterparts with identical memory footprint. Across a broad range of neural network architectures (deep CNNs, stacked LSTM, and seq2seq LSTM models), we find large-sparse models to consistently outperform small-dense models and achieve up to 10x reduction in number of non-zero parameters with minimal loss in accuracy.

Dynamic microscopy of nanoscale cluster growth at the solid–liquid interface

Abstract: Dynamic processes at the solid–liquid interface are of key importance across broad areas of science and technology. Electrochemical deposition of copper, for example, is used for metallization in integrated circuits, and a detailed understanding of nucleation, growth and coalescence is essential in optimizing the final microstructure. Our understanding of processes at the solid–vapour interface has advanced tremendously over the past decade due to the routine availability of real-time, high-resolution imaging techniques yielding data that can be compared quantitatively with theory1,2,3. However, the difficulty of studying the solid–liquid interface leaves our understanding of processes there less complete. Here we analyse dynamic observations—recorded in situ using a novel transmission electron microscopy technique—of the nucleation and growth of nanoscale copper clusters during electrodeposition. We follow in real time the evolution of individual clusters, and compare their development with simulations incorporating the basic physics of electrodeposition during the early stages of growth. The experimental technique developed here is applicable to a broad range of dynamic phenomena at the solid–liquid interface.

Hippocratic databases

Abstract: The Hippocratic Oath has guided the conduct of physicians for centuries. Inspired by its tenet of preserving privacy, we argue that future database systems must include responsibility for the privacy of data they manage as a founding tenet. We enunciate the key privacy principles for such Hippocratic database systems. We propose a strawman design for Hippocratic databases, identify the technical challenges and problems in designing such databases, and suggest some approaches that may lead to solutions. Our hope is that this paper will serve to catalyze a fruitful and exciting direction for future database research.

Designing for usability—key principles and what designers think

Abstract: Any system designed for people to use should be (a) easy to learn; (b) useful, i.e., contain functions people really need in their work; (c) easy to use; and (d) pleasant to use. In this note we present theoretical considerations and empirical data relevant to attaining these goals. First, we mention four principles for system design which we believe are necessary to attain these goals; Then we present survey results that demonstrate that our principles are not really all that obvious, but just seem obvious once presented. The responses of designers suggest they may sometimes think they are doing what we recommend when in fact they are not. This is consistent with the experience that systems designers do not often recommend or use them themselves. We contrast some of these responses with what we have in mind in order to provide a more useful description of our principles. Lastly, we consider why this might be so. These sections are summaries of those in a longer paper to appear elsewhere (Gould & Lewis, 1983). In that paper we elaborate on our four principles, showing how they form the basis for a general methodology of design, and we describe a successful example of using them in actual system design (IBM's Audio Distribution System).