Background For decades it has been assumed that postprimary tuberculosis is usually caused by reactivation of endogenous infection rather than by a new, exogenous infection. Methods We performed DNA fingerprinting with restriction-fragment–length polymorphism analysis on pairs of isolates of Mycobacterium tuberculosis from 16 compliant patients who had a relapse of pulmonary tuberculosis after curative treatment of postprimary tuberculosis. The patients lived in areas of South Africa where tuberculosis is endemic. Medical records were reviewed for clinical data. Results For 12 of the 16 patients, the restriction-fragment–length polymorphism banding patterns for the isolates obtained after the relapse were different from those for the isolates from the initial tuberculous disease. This finding indicates that reinfection was the cause of the recurrence of tuberculosis after curative treatment. Two patients had reinfections with a multidrug-resistant strain. All 15 patients who were tested for the human immu...

https://www.nejm.org/doi/pdf/10.1056/NEJM199910143411602

Exogenous Reinfection as a Cause of Recurrent Tuberculosis after Curative Treatment

In this study, the currently known typing methods for Mycobacterium tuberculosis isolates were evaluated with regard to reproducibility, discrimination, and specificity Therefore, 90 M tuberculosis complex strains, originating from 38 countries, were tested in five restriction fragment length polymorphism (RFLP) typing methods and in seven PCR-based assays In all methods, one or more repetitive DNA elements were targeted The strain typing and the DNA fingerprint analysis were performed in the laboratory most experienced in the respective method To examine intralaboratory reproducibility, blinded duplicate samples were included The specificities of the various methods were tested by inclusion of 10 non-M tuberculosis complex strains All five RFLP typing methods were highly reproducible The reliability of the PCR-based methods was highest for the mixed-linker PCR, followed by variable numbers of tandem repeat (VNTR) typing and spoligotyping In contrast, the double repetitive element PCR (DRE-PCR), IS6110 inverse PCR, IS6110 ampliprinting, and arbitrarily primed PCR (APPCR) typing were found to be poorly reproducible The 90 strains were best discriminated by IS6110 RFLP typing, yielding 84 different banding patterns, followed by mixed-linker PCR (81 patterns), APPCR (71 patterns), RFLP using the polymorphic GC-rich sequence as a probe (70 patterns), DRE-PCR (63 patterns), spoligotyping (61 patterns), and VNTR typing (56 patterns) We conclude that for epidemiological investigations, strain differentiation by IS6110 RFLP or mixed-linker PCR are the methods of choice A strong association was found between the results of different genetic markers, indicating a clonal population structure of M tuberculosis strains Several separate genotype families within the M tuberculosis complex could be recognized on the basis of the genetic markers used

Comparison of Methods Based on Different Molecular Epidemiological Markers for Typing of Mycobacterium tuberculosis Complex Strains: Interlaboratory Study of Discriminatory Power and Reproducibility

https://www.thelancet.com/pdfs/journals/lancet/PIIS0140-6736(01)06712-5.pdf

HIV-1 and recurrence, relapse, and reinfection of tuberculosis after cure: a cohort study in South African mineworkers

Molecular epidemiologic studies of tuberculosis (TB) have focused largely on utilizing molecular techniques to address short- and long-term epidemiologic questions, such as in outbreak investigations and in assessing the global dissemination of strains, respectively. This is done primarily by examining the extent of genetic diversity of clinical strains of Mycobacterium tuberculosis. When molecular methods are used in conjunction with classical epidemiology, their utility for TB control has been realized. For instance, molecular epidemiologic studies have added much-needed accuracy and precision in describing transmission dynamics, and they have facilitated investigation of previously unresolved issues, such as estimates of recent-versus-reactive disease and the extent of exogenous reinfection. In addition, there is mounting evidence to suggest that specific strains of M. tuberculosis belonging to discrete phylogenetic clusters (lineages) may differ in virulence, pathogenesis, and epidemiologic characteristics, all of which may significantly impact TB control and vaccine development strategies. Here, we review the current methods, concepts, and applications of molecular approaches used to better understand the epidemiology of TB.

Molecular Epidemiology of Tuberculosis: Current Insights

Conventional tests are not always helpful in making a diagnosis of tuberculous meningitis. We did a systematic review and meta-analysis to establish the summary accuracy of nucleic acid amplification (NAA) tests for tuberculous meningitis. We searched six electronic databases and contacted authors, experts, and manufacturers. Measures of diagnostic accuracy were pooled using a random effects model. 49 studies met our inclusion criteria. The summary estimates in 14 studies with commercial NAA tests were: sensitivity 0·56 (95% CI 0·46, 0·66), specificity 0·98 (0·97, 0·99), positive likelihood ratio 35·1 (19·0, 64·6), negative likelihood ratio 0·44 (0·33, 0·60), and diagnostic odds ratio 96·4 (42·8, 217·3). In the 35 studies with in-house (“homebrew”) tests, the summary accuracy could not be established with confidence because of wide variability in test accuracy. On current evidence, commercial NAA tests show a potential role in confirming tuberculous meningitis diagnosis, although their overall low sensitivity precludes the use of these tests to rule out tuberculous meningitis with certainty.

Diagnostic accuracy of nucleic acid amplification tests for tuberculous meningitis: a systematic review and meta-analysis

Large numbers of Mycobacterium tuberculosis isolates that were obtained from patients' sputa on diagnosis and during follow-up after short-course chemotherapy in Madras, India, have either no copy or only a single copy of IS6110. This poses a limitation for DNA fingerprinting with an IS6110-based probe to determine the frequency of exogenous reinfection versus that of endogenous reactivation. In the present study, we overcame this limitation by using an alternate probe, the direct-repeat element. Comparison of pre- and posttreatment isolates by direct-repeat restriction fragment length polymorphism analysis indicated a high degree of endogenous reactivation among patients who have relapses after the successful completion of chemotherapy.

Restriction fragment length polymorphism typing of clinical isolates of Mycobacterium tuberculosis from patients with pulmonary tuberculosis in Madras, India, by use of direct-repeat probe.

Solid state disks (SSDs) have the potential to revolutionize the storage system landscape, mostly due to their good random access performance, compared to hard disks. However, garbage collection (GC) in SSD introduces significant latencies and large performance variations, which renders widespread adoption of SSDs difficult. To address this issue, we present a novel garbage collection strategy, consisting of two components, called Advanced Garbage Collection (AGC) and Delayed Garbage Collection (DGC), that operate collectively to migrate GC operations from busy periods to idle periods. More specifically, AGC is employed to defer GC operations to idle periods in advance, based on the type of the idle periods and on-demand GC needs, whereas DGC complements AGC by handling the collections that could not be handled by AGC. Our comprehensive experimental analysis reveals that the proposed strategies provide stable SSD performance by significantly reducing GC overheads. Compared to the state-of-the-art GC strategies, P-FTL, L-FTL and H-FTL, our AGC+DGC scheme reduces GC overheads, on average, by about 66.7%, 96.7% and 98.2%, respectively.

/pdf/taking-garbage-collection-overheads-off-the-critical-path-in-52jjdb778s.pdf

Taking garbage collection overheads off the critical path in SSDs

Efficient resource management in data centers and clouds running large distributed data processing frameworks like MapReduce is crucial for enhancing the performance of hosted applications and increasing resource utilization. However, existing resource scheduling schemes in Hadoop MapReduce allocate resources at the granularity of fixed-size, static portions of nodes, called slots. In this work, we show that MapReduce jobs have widely varying demands for multiple resources, making the static and fixed-size slot-level resource allocation a poor choice both from the performance and resource utilization standpoints. Furthermore, lack of coordination in the management of multiple resources across nodes prevents dynamic slot reconfiguration, and leads to resource contention. Motivated by this, we propose MROrchestrator, a MapReduce resource Orchestrator framework, which can dynamically identify resource bottlenecks, and resolve them through fine-grained, coordinated, and on-demand resource allocations. We have implemented MROrchestrator on two 24-node native and virtualized Hadoop clusters. Experimental results with a suite of representative MapReduce benchmarks demonstrate up to 38% reduction in job completion times, and up to 25% increase in resource utilization. We further demonstrate the performance boost in existing resource managers like NGM and Mesos, when augmented with MROrchestrator.

MROrchestrator: A Fine-Grained Resource Orchestration Framework for MapReduce Clusters

Massively parallel scientific applications, running on extreme-scale supercomputers, produce hundreds of terabytes of data per run, driving the need for storage solutions to improve their I/O performance. Traditional parallel file systems (PFS) in high performance computing (HPC) systems are unable to keep up with such high data rates, creating a storage wall. In this work, we present a novel multi-tiered storage architecture comprising hybrid node-local resources to construct a dynamic data staging area for extreme-scale machines. Such a staging ground serves as an impedance matching device between applications and the PFS. Our solution combines diverse resources (e.g., DRAM, SSD) in such a way as to approach the performance of the fastest component technology and the cost of the least expensive one. We have developed an automated provisioning algorithm that aids in meeting the check pointing performance requirement of HPC applications, by using a least-cost storage configuration. We evaluate our approach using both an implementation on a large scale cluster and a simulation driven by six-years worth of Jaguar supercomputer job-logs, and show that our approach, by choosing an appropriate storage configuration, achieves 41.5% cost savings with only negligible impact on performance.

Provisioning a Multi-tiered Data Staging Area for Extreme-Scale Machines

In order to meet the increasing demands of present and upcoming data-intensive computer applications, there has been a major shift in the disk subsystem, which now consists of more disks with higher storage capacities and higher rotational speeds. These have made the disk subsystem a major consumer of power, making disk power management an important issue. People have considered the option of spinning down the disk during periods of idleness or serving the requests at lower rotational speeds when performance is not an issue. Accurately predicting future disk idle periods is crucial to such schemes. This paper presents a novel disk-idleness prediction mechanism based on Markov models and explains how this mechanism can be used in conjunction with a three-speed disk. Our experimental evaluation using a diverse set of workloads indicates that (i) prediction accuracies achieved by the proposed scheme are very good (87.5% on average); (ii) it generates significant energy savings over the traditional power-saving method of spinning down the disk when idle (35.5% onaverage); (iii) it performs better than a previously proposed multi-speed disk management scheme (19% on average); and (iv) the performance penalty is negligible (less than 1% on average). Overall, our implementation and experimental evaluation using both synthetic disk traces and traces extracted from real applications demonstrate the feasibility of a Markov-model-based approach to saving disk power.

Ramya Prabhakar

Papers

Restriction fragment length polymorphism typing of clinical isolates of Mycobacterium tuberculosis from patients with pulmonary tuberculosis in Madras, India, by use of direct-repeat probe.

Taking garbage collection overheads off the critical path in SSDs

MROrchestrator: A Fine-Grained Resource Orchestration Framework for MapReduce Clusters

Provisioning a Multi-tiered Data Staging Area for Extreme-Scale Machines

Markov Model Based Disk Power Management for Data Intensive Workloads