Search

Model and Serve is the final phase of a data analytics solution (Figure 8-1). In this phase, transformed data is consumed for the final output, either through visualization or any dependent applications. The entire data journey is planned, based on the target use case. In this chapter, the discussion is on the various scenarios that are applicable in this phase, and how to decide on technologies based on the cost and efficiency.

The data preparation and training phase is the most important phase of the data analytics solution. During this phase, data ingested from various sources is merged and crunched together (Figure 6-1). The transformed data further gets infused with machine learning models or is sent to the model and serve phase. The entire data journey is planned, based on the target use case. This phase has been split into two chapters. In this chapter, the discussion is on the various technologies that are applicable in this phase for data analytics. In the next chapter, there are in-depth discussions on advanced data analytics, data science, and various platforms available on Azure to accelerate this journey.

As they say, data is the new oil. This current decade has set a steep trajectory of data growth, and organizations have been greatly successful by harnessing the power of this data. Not only that, with the democratization of Artificial Intelligence and Machine Learning, building predictions has become easier. The infusion of AI/ML with data has given lots of advantages to plan future requirements or actions. Some of the classic use cases are customer churn analysis, fraud analytics, social media analytics, and predictive maintenance. In this chapter, the focus is on how the evolution of data happened and how it disrupted the data analytics space. Moreover, there will be a brief description of the concepts of relational and nonrelational data stores, enterprise data warehouse and big data systems.

Public cloud has been disrupting businesses in a big way. Data analytics is another area where a public cloud has brought lots of innovation. Organizations can build, scale, and consume these solutions with a faster pace and economical cost. Since there is a fair understanding of data lakes and data analytics basics by now, this chapter discusses the role of a public cloud to disrupt the market and to accelerate the adoption of data analytics solutions.

Data analytics solutions have become one of the most critical investments for organizations today. Organizations that are ahead of the curve have been spending lots of money, time, and resources on this practice. Major tech giants like ecommerce, social networking, and FMCG (fast-moving consumer goods) are heavily dependent on their data analytics solutions. A few examples of the outcomes of data analytics are customer 360-degree, real-time recommendations, fraud analytics, and predictive maintenance solutions. This chapter is designed to share an overview of the building blocks of data analytics solutions, based on our learning from large-scale enterprise projects and helping organizations to start this practice.

Chapter 6 was an in-depth discussion of Azure Kubernetes Service and its core concepts, including masters, agent nodes, application scaling, supported programming models, and the tools available to work with Visual Studio and so forth. AKS is a managed service that provides native capabilities to monitor an entire infrastructure. In this chapter, we will explore the various options to monitor the AKS ecosystem.

Previous chapters discussed the benefits and various challenges of microservices applications. By now, it’s quite clear that orchestrators are the backbone of the microservices ecosystem. We also discussed Azure Service Fabric and the various options to deploy microservices applications. Azure Kubernetes Service (AKS) is a vast subject; however, we will cover the fundamental details required to deploy microservices applications. There are practical scenarios along with hands-on code to give you a fair idea of how easy it is to adopt AKS.

In this chapter, we discuss the various critical factors of designing a database for microservices applications. As per the narrative of the microservices architecture, each microservice should have its own databases. Segregation based on data access helps fit the best technology to handle respective business problems. This means that a single application can use different database technologies, which is a concept called polyglot persistence. Let’s look at important factors to consider when building a microservices application, as well as how the power of the Microsoft Azure platform can be harnessed to build highly scalable, agile, and resilient solutions.

In the previous chapter, you learned the core concepts of Service Fabric, such as the application model, application scaling, supported programming models, and clusters. Service Fabric very effectively streamlines an application, cluster deployment, and the scaling of applications. In this chapter, you discover how you can effectively monitor a Service Fabric cluster and the applications deployed on it. We will create an ASP.NET Core-based application to demonstrate how easy it is to add application monitoring, and use Application Insights to troubleshoot issues in database calls and remote HTTP calls.

In the previous chapters, you learned about the evolution of microservices and the advantages and challenges of the microservices architecture. We also described various options for implementing microservices in Azure. In this chapter, we explain Azure Service Fabric’s offerings.

If we look back at how the data centers of organizations have evolved over the years, we realize a trend. Technological advancements in computing hardware and software are aimed at maximizing the use of the resources available. Today, we talk about hyperscale, with organizations adopting public cloud platforms to deliver all sorts of computing resources and cutting-edge technology platforms like IoT, blockchain, and machine learning services, and so forth. This enables organizations to shift their focus from operating and maintaining their data centers to addressing their business challenges using a public cloud.

This is an era of digital transformation. Easy access of the Internet has empowered people and organizations to achieve more from sitting anywhere in the world. There was a time when every company was rushing for a web presence, and Internet adoption was picking up speed. Today, the Internet is in every pocket: you can access the Internet anywhere through cell phones, laptops, tablets, or PCs. This has brought a radical change to every industry. Today, every organization wants to do business online—whether retail, finance, entertainment, gaming, or so forth.

Because physical inactivity in chronic obstructive pulmonary disease (COPD) predicts health care use and mortality, we prospectively evaluated the relationship of this variable to the frequency of 30-day readmissions after a hospitalization for an exacerbation.Consented patients discharged after an exacerbation of COPD were asked to wear a GT3X+ accelerometer (ActiGraph, Pensacola, FL) continuously on the wrist for 30 days after hospital discharge. Vector magnitude units (VMU), the sum of movements in three planes over each minute of use, were recorded. Higher physical activity for each minute was defined by a VMU threshold of at least 3,000 counts. Those patients with fewer than 60 minutes/day over the first week were considered inactive.Fifty-four study patients were discharged from the hospital and 38 underwent activity testing. In the latter subgroup, all-cause hospital readmissions within 30 days occurred in 12 patients (32%). Minutes of higher physical activity per day over the first week after discharge were considerably lower in those eventually readmitted than in those who remained as outpatients: 42 ± 14 (SE) versus 114 ± 19 minutes, respectively (P = 0.02). In addition, physical activity decreased over time in those who were eventually readmitted, but increased in those who were not readmitted. Those with lower physical activity over Week 1 after discharge were more likely to have 30-day all-cause readmissions than those with higher activity: odds ratio, 6.7, P = 0.02. In multivariate testing, both physical inactivity and a history of two or more hospitalizations for exacerbations in the preceding year predicted 30-day readmission.These findings underscore the importance of physical activity as a predictor of this type of health care use outcome.

Summary: Background: Prevention of tuberculosis (TB) in the United States usually involves testing for latent tuberculosis infection (LTBI) with a tuberculin skin test (TST), followed by offering therapy to those who have a positive test result. QuantiFERON-TB Gold assay (QFT-G) is more specific for infection with Mycobacterium tuberculosis than the TST, especially among persons vaccinated with bacillus Calmette-Guérin, thereby reducing the number of false positive tests. Methods: Adults referred to a pulmonary clinic for a positive TST result were tested with QFT-G. We assessed factors for having a positive QFT-G. Results: Among 100 adults who were BCG-vaccinated and had a positive TST result, 30 (30%) had a positive result using QFT-G. Persons from high-incidence countries were 8.2 times more likely to have a positive QFT-G result compared with persons from low-incidence countries (46% versus 9%). Using logistic regression to assess QFT-G positivity, strong predictors included having an abnormal chest radiograph consistent with healed TB, a TST induration of ≥16 mm, and birth in a high-incidence country. Conclusion: Use of QFT-G assay following a positive TST result further identifies persons who would most benefit from treatment for LTBI. Keywords: Tuberculosis, Latent tuberculosis infection, Diagnosis, Treatment, Prevention

Background: Lower physical activity in COPD predicts increased health care utilization. Activity after a CHF exacerbation may be low and may also predict health care utilization. Objective: Compare post-discharge physical activity in CHF and COPD. Methods: We measured activity over 1 month (using tri-axial accelerometers on the wrist) in 37 discharged after a hospitalization for CHF. Minutes per day above 3000 vector magnitude units (sum of movements per minute in 3 planes) from the wrist device were analyzed; this corresponds to light-moderate intensity. Physical activity was compared with that from 38 discharged COPD patients from a similarly-designed earlier study. We then evaluated activity as a predictor of 30-day, all-cause hospital re-admissions in both diseases. Results: Re-admissions in CHF and COPD were 32 and 24%, respectively. Overall, higher-intensity activity in CHF and COPD were similar over 4 weeks, and, in both, physical activity was lower in those eventually re-admitted (Figure 1). Lower activity tended to predict readmission in both diseases: odds ratio (CHF) = 5.0 and (COPD) = 6.7 (p = 0.05 and 0.02, respectively). Conclusions: Physical activity post-discharge is similar in CHF and COPD, and predicts 30-day all-cause hospital re-admissions in both. At present, we do not know how these activity levels compare to those of healthy controls.