Big Data Architect Flashcards

Millions of people use a large telecommunications provider. Most of their clients pay in advance. They can very quickly move to other vendors because they are prepaid clients. This business has experienced some significant customer loss due to competition over the last four to six months. They want to create a system that can provide them access to the social network of their clients (e.g. who is the influencer and who is the follower). Additionally, they want the system to be educated over time to anticipate potential complaints and the capability to analyze voice and data consumption patterns in real time. Which of the following would you advise in this situation?

Answer: Spark

Apache Spark is a powerful big data processing engine that provides fast and distributed data processing capabilities. It is designed to handle large-scale data analytics and is well-suited for real-time data processing, machine learning, and stream processing.

A telecommunications company has a high rate of customer turnover. More than five million people use them. Each year, they have more than 400 terabytes of call detail records. Since the majority of their clients are prepaid, they are always free to switch telecom companies. They seek to comprehend client behavior in order to correct the situation. They intend to create a unique profile for each customer as a result. Additionally, they want to add social media information to the profile to enhance it. Which of the following would you advise, given these conditions?

Answer: Hadoop

Hadoop is an open-source big data processing framework that excels at handling large volumes of data, making it well-suited for analyzing vast amounts of call detail records and customer data.

Which of the following claims about cloud applications is TRUE?

Answer: Leveraging a private vs. public cloud may result in sacrificing some of the core advantages of cloud computing

The statement that is TRUE regarding cloud applications is Leveraging a private vs. public cloud may result in sacrificing some of the core advantages of cloud computing. It's essential for organizations to carefully consider their requirements, workload characteristics, and cost considerations before deciding between private and public clouds. Each deployment model has its advantages and trade-offs, and the choice should align with the organization's specific needs and business objectives.

Different degrees of Service Level Agreements (SLAs) definition exist. Which of the following is NOT a valid level?

Answer: Multilevel SLA

Service Level Agreements (SLAs) are contracts or agreements between a service provider and its customers that define the expected level of service and the metrics that will be used to measure the performance of the service. SLAs can be defined at different levels, but "Multilevel SLA" is not a recognized or standard term.

What example of unstructured data is NOT one of the ones listed below?

Answer: Netezza table

Netezza table is a data warehouse appliance that uses a columnar storage format. It stores structured data in columns and rows, making it a structured data storage solution.

What does the term "NoSQL" actually mean?

Answer: It is not limited to relational database technology

The effective meaning of "NoSQL" is: It is not limited to relational database technology. NoSQL stands for "Not Only SQL" or "Non-Relational," and it refers to a class of database management systems that do not strictly adhere to the traditional relational database model. NoSQL databases provide an alternative approach to storing and retrieving data, and they are designed to handle large volumes of unstructured, semi-structured, or structured data more efficiently than traditional relational databases.

Data in motion is information that is continuously being added to. Which of the following can be used to import this kind of data into the distributed file system?

Answer: Flume

Flume is a distributed data collection service provided by the Apache Hadoop ecosystem. It is designed to efficiently collect, aggregate, and move large amounts of streaming data (data in motion) from various sources into Hadoop's distributed file system (HDFS) for further processing and analysis. Flume supports a wide range of data sources, including log files, social media feeds, sensors, and more.

Which of the following big data elements decides whether to replicate blocks at all?

Answer: Name Node

The NameNode in HDFS (Hadoop Distributed File System) is the master node responsible for managing the file system namespace and metadata. It makes all decisions regarding block replication, including the replication factor for data blocks and instructing DataNodes to replicate or delete blocks. DataNodes merely store the data blocks and report their status to the NameNode.

Which of the following describes network congestion evidence?

Answer: Packet Discards

Network congestion occurs when the volume of traffic exceeds the network's capacity, leading to performance degradation. Packet discards, or dropped packets, are a direct and undeniable symptom of congestion, as network devices discard packets when their buffers become full. While other options like user complaints or traffic volumes can indicate potential issues, packet discards are a definitive technical evidence of congestion.

What task must be completed to achieve the service level requirement (SLR), which is fewer than 3 milliseconds?

Answer: Measure switch failure frequency

To achieve a stringent Service Level Requirement (SLR) of less than 3 milliseconds, it's crucial to eliminate sources of significant latency and downtime. Switch failures introduce considerable delays and outages, directly impacting network performance and potentially exceeding the strict latency target. Measuring switch failure frequency helps identify unreliable hardware that could cause service interruptions, allowing for proactive maintenance or replacement to maintain the desired low latency.

Which of the following best describes a quality criteria or constraint that a system (or a specific component of a system) must meet?

Answer: Service Level Agreement

A Service Level Agreement (SLA) is a formal contract that defines the specific quality criteria and constraints a system or service must meet. It translates abstract quality attributes, often referred to as non-functional requirements, into measurable and enforceable terms like uptime, response times, and performance targets. Therefore, an SLA best describes the formalized and agreed-upon quality criteria that a system is contractually obligated to satisfy.

What is TRUE about the following assertions about SPSS?

Answer: SPSS software provides a security framework

IBM SPSS Statistics, as an enterprise-grade analytical software, is equipped with a comprehensive security framework. This framework is essential for protecting sensitive data and controlling access to analytical models and results. It includes features for user authentication, authorization, and data encryption, ensuring data confidentiality, integrity, and availability within the SPSS environment.

Which of the Big SQL-related statements below is TRUE?

Answer: Big SQL supports updates in Hive.

IBM Big SQL extends the capabilities of Hive by providing a full SQL interface to Hadoop data, including support for Data Manipulation Language (DML) operations. Unlike standard Hive, which traditionally has limited or no direct support for updates, Big SQL allows users to perform `UPDATE`, `INSERT`, and `DELETE` statements directly on Hive tables. This feature significantly enhances the flexibility and utility of Hive for transactional workloads within the Hadoop ecosystem.

A bank wants to develop a system that keeps track of all real-time internet and ATM transactions. They intend to use both enterprise and social media data to create a customized model of their consumers' financial activity. Over time, the system must be able to learn and adjust. These customized models will be utilized for in-the-moment advertising as well as for the identification of any fraud or criminal activity. Which of the following recommendations makes sense in light of given conditions?

Answer: Spark

The scenario describes a need for real-time processing of large transaction volumes, integration of diverse data sources, and the ability to build adaptive, learning models for fraud detection and personalized advertising. Apache Spark is an ideal solution due to its in-memory processing capabilities for high-speed real-time analytics and its integrated machine learning library (MLlib). Its streaming capabilities further enable continuous learning and immediate action on fast-moving data streams.

Which of the following objectives does BigInsights support?

Answer: Supports data exchange with a number of sources

IBM BigInsights is an enterprise-grade Hadoop distribution designed to integrate seamlessly with existing IT infrastructures. A primary objective of BigInsights is to support robust data exchange and integration with a wide array of data sources. This includes traditional relational databases, data warehouses, streaming data platforms, and other enterprise applications, enabling organizations to leverage their diverse data assets within the Hadoop ecosystem for comprehensive analytics.

To analyze client sales data and forecast which products will sell better, you must set up a Hadoop cluster. Which of the following options will allow you to build up your cluster with the highest platform stability?

Answer: Leverage the Open Data Platform (ODP) core to provide a stable base against which Big Data solutionsproviders can qualify solutions

To achieve the highest platform stability for a Hadoop cluster, leveraging the Open Data Platform (ODP) core is the recommended approach. ODP provides a standardized, stable, and tested set of core Hadoop components, ensuring interoperability and reliability across different vendors and solutions. Building on this consistent foundation minimizes integration challenges and enhances overall system stability, allowing solution providers to qualify their offerings against a reliable base.