MapReduce has been widely accepted as a simple programming pattern that can form the basis for efficient, large-scale, distributed data processing. The success of the MapReduce pattern has led to a variety of implementations for different computational scenarios. In this paper we present MRJ, a MapReduce Java framework for multi-core architectures. We evaluate its scalability on a fourcore, hyperthreaded Intel Core i7 processor, using a set of standard MapReduce benchmarks. We investigate the significant impact that Java runtime garbage collection has on the performance and scalability of MRJ. We propose the use of memory management autotuning techniques based on machine learning. With our auto-tuning approach, we are able to achieve MRJ performance within 10% of optimal on 75% of our benchmark tests.
We report the case of a 57-year-old woman presented to our institute with complaints of progressive imbalance while walking, double vision on looking to right lateral aspect, deviation of angle of the mouth toward the left side, inability to close right eyelids completely, and speaking difficulty of 2 weeks duration. Her clinical examination showed horizontal nystagmus with fast component towards right side, right abducens nerve palsy, right lower motor neuron (LMN) facial nerve palsy, staccato speech, and right-side appendicular ataxia. Her erythrocyte sedimentation rate (ESR) was 57 mm/h and chest X-ray was normal. Magnetic resonance (MR) images revealed a large thick irregularwalled ring-enhancing lesion in the right hemipons (►Fig. 1A-C). MR spectroscopy from the lesion revealed large lipid peak without significant choline elevation. Findings were consistent with inflammatory etiology, possibly tuberculoma.The patient was started on antituberculosis medications and periodic follow-up MR images were obtained. The initial thick-walled intensely enhancing ring lesion became thin Keywords ► central nervous system tuberculosis ► brainstem tuberculosis ► tuberculoma ► magnetic resonance imaging AbstractIncidence of tuberculosis (TB) is decreasing all over the world due to improved diagnosis and treatment. Though central nervous system TB accounts for 1% of all TB cases, it carries high mortality and neurologic morbidity. Surgery is difficult when eloquent areas of brain are involved, particularly brainstem. Even with advent of magnetic resonance imaging (MRI) facilities and serologic tests, definitive diagnosis is difficult. We report a case of a 57-year-old-woman with possible brainstem TB, diagnosed based on initial MR images. She was empirically started on anti-TB medications followed by serial repeat MR images at 5, 12, and 19 months interval. The initial MR images revealed a thick ring-enhancing lesion, which showed progressive reduction in size, became irregular thin-walled ring-enhancing lesion at 5 months, uniform ring-enhancing lesion at 12 months, and small solid nodular enhancement at 19 months follow-up. There was gradual improvement in neurologic status, and the patient became asymptomatic at the end of 1 year. Characteristic MRI changes over a period favoring adequate response to antituberculosis treatment aids in strengthening presumptive diagnosis of brainstem tuberculoma in resource-poor countries. Along with presumptive diagnosis and timely initiation of medical therapy, periodic MRI follow-up is important in preventing mortality and reducing morbidity.
The transsphenoidal approach to lesions of the sella turcica, especially for pituitary tumors, has been favored by neurosurgeons because of decreased perioperative patient morbidity and mortality of this approach compared with others. It is widely accepted because of its minimally invasive and maximally effective procedure. We report a rare case of a 63-year-old female patient with the asymptomatic delayed occurrence of sphenoid pyocele, 6 years after the initial pituitary surgery, and remained asymptomatic during this long interval.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.