Browsing by Subject "compression"
Now showing 1 - 7 of 7
Results Per Page
Sort Options
Item An energy efficient TCAM enhanced cache architecture(Texas A&M University, 2005-08-29) Surprise, Jason MathewMicroprocessors are used in a variety of systems ranging from high-performance super computers running scientific applications to battery powered cell phones performing realtime tasks. Due to the large disparity between processor clock speed and main memory access time, most modern processors include several caches, which consume more than half of the total chip area and power budget. As the performance gap between processors and memory has increased, the trend has been to increase the size of the on-chip caches. However, increasing the cache size also increases its access time and energy consumptions. This growing power dissipation problem is making traditional cooling and packaging techniques less effective thus requiring cache designers to focus more on architectural level energy efficiency than performance alone. The goal of this thesis is to propose a new cache architecture and to evaluate its efficiency in terms of miss rate, system performance, energy consumption, and area overhead. The proposed architecture employs the use of a few Ternary-CAM (TCAM) cells in the tag array to enable dynamic compression of tag entries containing contiguous values. By dynamically compressing tag entries, the number of entries in the tag array can be reduced by 2N, where N is the number of tag bits that can be compressed. The architecture described in this thesis is applicable to any cache structure that uses Content Addressable Memory (CAM) cells to store tag bits. To evaluate the effectiveness of the TCAM Enhanced Cache Architecture for a wide scope of applications, two case studies were performed ?? the L2 Data-TLB (DTLB) of a high-performance processor and the L1 instruction and data caches of a low-power embedded processor. Results indicate that a L2 DTLB implementing 3-bit tag compression can achieve 93% of the performance of a conventional L2 DTLB of the same size while reducing the on-chip energy consumption by 74% and the total area by 50%. Similarly, an embedded processor cache implementing 2-bit tag compression achieves 99% of the performance of a conventional cache while reducing the on-chip energy consumption by 33% and the total area by 10%.Item Efficient Algorithms for Comparing, Storing, and Sharing Large Collections of Phylogenetic Trees(2012-07-16) Matthews, SuzanneEvolutionary relationships between a group of organisms are commonly summarized in a phylogenetic (or evolutionary) tree. The goal of phylogenetic inference is to infer the best tree structure that represents the relationships between a group of organisms, given a set of observations (e.g. molecular sequences). However, popular heuristics for inferring phylogenies output tens to hundreds of thousands of equally weighted candidate trees. Biologists summarize these trees into a single structure called the consensus tree. The central assumption is that the information discarded has less value than the information retained. But, what if this assumption is not true? In this dissertation, we demonstrate the value of retaining and studying tree collections. We also conduct an extensive literature search that highlights the rapid growth of trees produced by phylogenetic analysis. Thus, high performance algorithms are needed to accommodate this increasing production of data. We created several efficient algorithms that allow biologists to easily compare, store and share tree collections over tens to hundreds of thousands of phylogenetic trees. Universal hashing is central to all these approaches, allowing us to quickly identify the shared evolutionary relationships contained in tree collections. Our algorithms MrsRF and Phlash are the fastest in the field for comparing large collections of trees. Our algorithm TreeZip is the most efficient way to store large tree collections. Lastly, we developed Noria, a novel version control system that allows biologists to seamlessly manage and share their phylogenetic analyses. Our work has far-reaching implications for both the biological and computer science communities. We tested our algorithms on four large biological datasets, each consisting of 20; 000 to 150; 000 trees over 150 to 525 taxa. Our experimental results on these datasets indicate the long-term applicability of our algorithms to modern phylogenetic analysis, and underscore their ability to help scientists easily exchange and analyze their large tree collections. In addition to contributing to the reproducibility of phylogenetic analysis, our work enables the creation of test beds for improving phylogenetic heuristics and applications. Lastly, our data structures and algorithms can be applied to managing other tree-like data (e.g. XML).Item Efficient Parallel Text Compression on GPUs(2012-02-14) Zhang, XiaoxiThis paper demonstrates an efficient text compressor with parallel Lempel-Ziv-Markov chain algorithm (LZMA) on graphics processing units (GPUs). We divide LZMA into two parts, match finder and range encoder. We parallel both parts and achieve competitive performance with freeArc on AMD 6-core 2.81 GHz CPU. We measure match finder time, range encoder compression time and demonstrate realtime performance on a large dataset: 10 GB web pages crawled by IRLbot. Our parallel range encoder is 15 times faster than sequential algorithm (FastAC) with static model.Item Evaluation of Basic Parameters for Packaging, Storage and Transportation of Biomass Material from Field to Biorefinery(2012-02-14) Paliwal, RichaThe universal adoption of biomass materials as an alternate fuel source to fossil fuels for transportation and electricity has been hindered by the high transportation costs involved in fuel production. Optimization of these initial costs will make the eco-friendly fuels more economically viable. Biomass is a promising feedstock for biofuels primarily because it is a renewable and sustainable resource. Among the most studied grassland crops, switchgrass is a perennial warm-season grass and has been identified as a potential energy crop. This research focuses on evaluating various physical parameters which affect the economic feasibility of packaging and transporting switchgrass from the field to the biorefinery. The switchgrass was harvested using a mower conditioner followed by field chopping after varying drying periods. The first harvesting period spanned from early November to mid December 2007 and the second was August to October 2008. Densification properties of chopped switchgrass were studied under compression. The effects of compressive stresses (41 to 101 kPa), number of strokes (1 to 10), moisture content (9 to 62 percent) and chopping length (63 and 95 mm) on the densification of chopped switchgrass were studied. The final dry matter density (DMD) increased with the compressive stresses and the number of strokes, small chop length and low moisture content. The maximum free-standing DMD obtained was 245 kg/m^3.Item Experimental evaluation of wire mesh for design as a bearing damper(Texas A&M University, 2004-11-15) Choudhry, Vivek VaibhavWire mesh vibration dampers have been the subject of some very encouraging experiments at the Texas A&M Turbomachinery laboratories for the past several years and have emerged as an excellent replacement for squeeze film dampers. Their capability to provide damping for a wide range of temperatures (even cryogenic), fluid free operation and ability to perform even when soaked with lubricants makes them a suitable option as a bearing damper. Experiments were conducted to investigate the effect of design parameters like axial thickness and axial compression that influence the characteristics of wire mesh as a bearing damper. Two groups of wire mesh were tested to show that the stiffness and damping are directly proportional to the axial thickness, if all the other parameters are kept constant. Tests on four wire mesh donuts of different radial thickness showed that stiffness and damping vary inversely with radial thickness. Rigorous tests were also conducted to quantify the effects of axial compression, radial interference and displacement amplitude on stiffness and damping of the wire mesh. Another novel kind of mesh damper tested was comprised of two small segments instead of a whole donut. The results showed that wire mesh exhibited good damping characteristics even when used in small segments. Empirical expressions were developed using MathCADTM worksheets, and an existing ExcelTM design worksheet was modified to include these factors. The effect of frequency variation was also included to give a comprehensive design tool for wire mesh. A new design worksheet was developed that can predict rotordynamic coefficients for a wire mesh bearing damper having a different size as well as different installation and operational conditions.Item Overpressure prediction by mean total stress estimate using well logs for compressional environments with strike-slip or reverse faulting stress state(Texas A&M University, 2007-04-25) Ozkale, AslihanPredicting correct pore-pressure is important for drilling applications. Wellbore stability problems, kicks, or even blow-outs can be avoided with a good estimate of porepressure. Conventional pore-pressure estimation methods are based on one-dimensional compaction theory and depend on a relationship between porosity and vertical effective stress. Strike-slip or reverse faulting environments especially require a different way to determine pore-pressure, since the overburden is not the maximum stress. This study proposes a method which better accounts for the three-dimensional nature of the stress field and provides improved estimates of pore-pressure. We apply the mean total stress estimate to estimate pore-pressure. Pore pressure is then obtained by modifying Eaton??????s pore-pressure equations, which require either resistivity or sonic log data. The method was tested in the Snorre Field in the Norwegian North Sea, where the field changes from strike-slip to reverse stress state. Eaton??????s resistivity and sonic equations were used to predict pore-pressure in this region by replacing the vertical stress by the mean total stress estimate. Results suggest that the modified Eaton method with resistivity log data gives better results for the area than the conventional method. The ratio of maximum horizontal stress to minimum horizontal stress throughout each well should be known for best results.Item Static Pressure Loss in 12?, 14?, and 16? Non-metallic Flexible Duct(2013-08-01) Cantrill, David LeeThis study was conducted to determine the effects of compression on pressure drops in non-metallic flexible duct. Duct sizes of 12?, 14? and 16? diameters were tested at a five different compression ratios (maximum stretch, 4%, 15%, 30% and 45%) following the draw through methodology in ASHRAE Standard 120 -1999 ? Methods of Testing to Determine Flow Resistance of Air Ducts and Fittings. With the pressure drop data gathered, equations were developed to approximate the pressure loss at a given air flow rate for a given duct size. The data gathered showed general agreement with previous studies showing an increase in compression ratio leads to an increase in static pressure loss through the duct. It was determined that pressure losses for compression ratios greater than 4% were over four times greater than maximum stretched flexible duct of corresponding duct size. The increased static pressure losses can lead to decreased performance in HVAC systems. The findings of this study add to the existing ASHRAE and industry data for flexible duct with varying compression ratios.