DataOps Methodology Cognitive Class Exam Answers:-

Machine Learning January 22, 2022 cognitive class Leave a comment 2,346 Views

Module 1 :- Establish DataOps

Lesson 2 : Establish Data Strategy

Question 1 : Before we can put together a data strategy, we need to have a good understanding of the data available and how it is used in the organization.

True
False

Question 2 : What is a data strategy?

An architecture and actionable roadmap along with an action plan
A competitive publication to show that our organization is modern
A plan to move all legacy data systems to the cloud

Question2 : Implementing a data strategy should always result in cost savings in the year the plan is realized.

True
False

Question3 : Which of the following statements about Data Strategy are correct?

Whatever the type of data, it should only include internally produced data
All types of data – both structured and unstructured need to be considered
Volumes of data have increased hugely, but are now starting to stabilize
Only business executives should be consulted in putting together a strategy

Question 4 : Data Governance is a key part of executing a data strategy.

True
False

Lesson 3 :- Establish Team

Question 1 : A DataOps team consists of members mostly from IT departments.

True
False

Question 2 : Which of the following roles are active team members of any DataOps team?

Chief Technology Officer
Chief Data Officer
Data Engineer
Database Administrator
Data Steward
Data Architect
Data Scientist

Question 3 : Creating and maintain business terms is a major responsibility of which following role?

Data Engineer
Data Quality Analyst
Data Steward
Data Scientist

Question 4 : Only Chief Data Officer can update the KPIs for a data sprint.

True
False

Question 5 : DataOps relies heavily on the use of automation, so that communication among team members is not necessary.

True
False

Module 2 :- Establish DataOps-Optimize for operation

Lesson 1 :-Establish Toolchain

Question 1: DataOpstoolchain helps you deliver quality data slowly.

True
False

Question 2 : DataOpsToolchain and DevOps are the same thing.

True
False

Question 3 : DataOpsToolchain can work without DataOps API(s).

True
False

Question 4 : What are the key components of DataOpsToolchain?

Continuous Deployment
Communication
Source Control
All of above

Question 5 : Who is responsible for creating DataOpsToolchain? (Choose all that apply)

Data Scientist
Administrator
DBA
Data Engineer

Lesson 2 :- Establish Baseline

Question 1 : Data Management is the same as Information Governance.

True
False

Question 2 : What is the most costly result from an external influence to an organization?

Data Breach Fines and Penalties
Insurance Policy Payout
Claim Settlement
None of these

Question 3 : Reference data is defined as data used as a permissible value within a data field.

True
False

Lesson 3 :- Establish Business priorities

Question 1 : Business Priority should be the primary focus when deciding what the DataOps team should do.

True
False

Question 2 : What is a data backlog?

A bottleneck in the data pipeline
A list of all data sources
A prioritized set of requirements expressed as data tasks
A plan to move all data into a catalog

Question 3 : A prioritized data backlog will reduce the time taken to start the next DataOps iteration.

True
False

Question 4 : A Data Task should be prioritized by considering:

The cost of providing the data
The career advancement possibilities of solving business challenges
The impact to sales from implementing the data pipeline
All of the above

Question 5: KPIs are used to determine the progress and throughput of a DataOps data sprint.

True
False

Module 3. Iterate Dataops – Know Your Data

Lesson 1. Discover

Question 1. You will need someone on your team with detailed knowledge of the business processes you’re going to analyze so selected data elements are appropriate to reaching your objectives.

True
False

Question 2. What should you do if you identify gaps or mismatches in the data required for the analysis?

Rethink how you will do the analysis with different data
Create the missing data
Find a new source for the missing or mismatched data
All of the above

Question 3. You should trace the linage of data elements to be used for analysis to make sure they come from a trusted source.

True
False

Question 4. What is the primary objective of the Discover phase?

Decide what the analytics team wants to have for lunch
Identify and locate the specific data elements required to accomplish an analysis
Uncover the meaning of data column headers and how they relate to the underlying data
Gain an understanding of the business goals and KPIs of an analysis effort

Question 5. A Data Engineer who thoroughly understands where specific data resides, including the specific databases and files where each identified data element resides, should be involved in Data Discovery process.

True
False

Lesson 2. classify

Question 1. Classification of each data element will make it easier going forward for users to distinguish the meaning and applicability of the data for their purposes.

True
False

Question 2. Which description best defines taxonomy?

Organizing data elements into meaningful structures
An IBM network protocol which reduces network latency
The art of preparing, stuffing, and mounting the skins of animals with lifelike effect

Question 3. A single data element can be placed into an unlimited number of data domains.

True
False

Question 4. Which of the following is the objective of classification?

To bring out points of similarity and dissimilarity among various groups
To present data in a simple, logical and understandable form
To condense the mass of data
All of the above

Question 5. You should design workflows which are specific to the classification tool you are using.

True
False

Module 4. Iterate Dataops

Lesson 1. Manage Qualities & Entities

Question 1. Data quality is data accuracy.

True
False

Question 2. All data across the enterprise should have the same data quality.

True
False

Question 3. A data quality framework consists of which of the following 4 phases:

Profile
Define
Remediate
Monitor
Assess
Deploy

Question 4. When assessing data quality, you only need the data set containing the data, metadata is optional.

True
False

Lesson 2. Manage Policies

Question 1. How does data classification affect defining policies?

Inheritance, retention and probabilities
Protection, reporting and inheritance
Protection, accessibility and retention
Retention, deletion and storage

Question 2. What impact does a highly sensitive classification have on a policy definition?

Require data anonymization, de-identification, and masking
Limit access to the data and/or require data masking
Limit access to the data and make it unprintable
No impact

Question 3.What are the most common state, country or regional regulations affecting personal information?

SIN, SSN and BAN
FDIC, BCBS and SOX
CCPA, GDPR and LGPD
PCI, PII and PHI

Question 4. Once policies have been defined affecting the data, rules must be enforced to act.

True
False

Module 5. Iterate Dataops – Use Your Data

Lesson 1. Self Service

Question 1. Self Service of data is only possible when any data movement and transformation required to join multiple data assets have been performed.

True
False

Question 2. Self Service can use the following governance artefacts to refine a search in a catalog. (Choose all that apply)

Data Protection Rules
Business Terms
Tags

Question 3. A data consumer should not be able to access data that has been identified as sensitive, where there is not a business need to do so.

True
False

Question 4. Which of the following statements about Self Service are correct?

Data consumers typically do not know how to manipulate the data
Data Protection rules prevent a data consumer from inadvertently seeing data that is sensitive
Creating multiple catalogs can partition data assets by their content and anticipated audience
A data consumer needs to know SQL to join multiple data assets

Question 5. Data Consumers provide valuable input to data scientists by clarifying the combination of data assets and how they need to be transformed, prior to data movement being designed and implemented.

True
False

Lesson 2. Manage Movement & Integration

Question : 1 You should define the use case at the outset of a Data Movement and Integration project to support a “Build It and They Will Come” strategy.

True
False

Question 2. Which of the following does not represent a data integration pattern:

Data virtualization
Data replication
Data lineage
Message-oriented movement
Bulk/batch

Question 3. Which of the following is not a Data Movement and Integration Job Design consideration?

Design for reusability
Deployment models (e.g. Containers, Kubernetes Orchestration, OpenShift)
Design for parallel processing
Everything should be programmed in Python
Design for job portability (build once and run anywhere)

Question 4. Hand coding generally provides a 10X productivity gain over commercial data integration software tooling.

True
False

Question 5. Which of the following is not an example of a message queuing system?

Kafka
VSAM
Microsoft Azure Queues
GCP PubSub
AWS Simple Queue Service
MQ

Module 6. Improve Dataops

Question 1. DataOps is a fixed process which should not be changed once defined.

True
False

Question 2. Improvements to the DataOps process could involve changes to

Technology used in DataOps
DataOps team roles and responsibilities
Processes for ETL
All of the above

Question 3. Reviewing the Data classification phase involves reviewing how accurate the data mappings to the business terms are.

True
False

Question 4. Reviewing the Establish Baseline Process should include reviewing how effective the processes are for establishing a baseline for –

External Regulatory requirements
Organization maturity and Readiness
Governance and Oversight
All of the above

Question 5. KPIs are key in determining the effectiveness of all parts of the DataOps process.

True
False

DataOps Methodology Cognitive Class Final Exam Answers:-

Question 1. What is a data strategy?

An architecture and actionable roadmap along with an action plan
A competitive publication to show that our organization is modern
A plan to move all legacy data systems to the cloud

Question 2. Which of the following statements about Data Strategy are correct?

Whatever the type of data, it should only include internally produced data
All types of data – both structured and unstructured need to be considered
Volumes of data have increased hugely, but are now starting to stabilize
Only business executives should be consulted in putting together a strategy

Question 3. Which of the following roles are active team members of any DataOps team?

Chief Technology Officer
Chief Data Officer
Data Engineer
Database Administrator
Data Steward
Data Architect
Data Scientist

Question 4. Creating and maintaining business terms is a major responsibility of which following role?

Data Engineer
Data Quality Analyst
Data Steward
Data Scientist

Question 5. Business Priority should be the primary focus when deciding what the DataOps team should do.

True
False

Question 6.What is a data backlog?

A bottleneck in the data pipeline
A list of all data sources
A prioritized set of requirements expressed as data tasks
A plan to move all data into a catalog

Question 7.A Data Task should be prioritized by considering:

The cost of providing the data
The career advancement possibilities of solving business challenges
The impact to sales from implementing the data pipeline
All of the above

Question 8. KPIs are used to determine the progress and throughput of a DataOps data sprint.

True
False

Question 9.What are key components of DataOps toolchain?

Continuous Deployment
Communication
Source Control
All of above

Question 10.Who is responsible for creating DataOps toolchain? (Choose all that apply)

Data Scientist
Administrator
DBA
Data Engineer

Question 11.What is the primary objective of the Discover phase?

Decide what the analytics team wants to have for lunch.
Identify and locate the specific data elements required to accomplish an analysis
Uncover the meaning of data column headers and how they relate to the underlying data.
Gain an understanding of the business goals and KPIs of an analysis effort.

Question 12.Which description best defines taxonomy?

Organizing data elements into meaningful structures.
An IBM network protocol which reduces network latency.
The art of preparing, stuffing, and mounting the skins of animals with lifelike effect.

Question 13. Which of the following is the objective of classification?

To bring out points of similarity and dissimilarity among various groups.
To present data in a simple, logical and understandable form.
To condense the mass of data.
All of the above

Question 14. A data quality framework consists of which of the following 4 phases:

Profile
Define
Remediate
Monitor
Assess
Deploy

Question 15.How does data classification affect defining policies?

Inheritance, retention and probabilities
Protection, reporting and inheritance
Protection, accessibility and retention
Retention, deletion and storage

Question 16. What impact does a highly sensitive classification have on a policy definition?

Require data anonymization, de-identification, and masking
Limit access to the data and/or require data masking
Limit access to the data and make it unprintable
No impact

Question 17. Self Service can use the following governance artefacts to refine a search in a catalog. (Choose all that apply)

Data Protection Rules
Business Terms
Tags

Question 18. Which of the following statements about Self Service are correct?

A data consumer needs to know SQL to join multiple data assets
Data Protection rules prevent a data consumer from inadvertently seeing data that is sensitive
Creating multiple catalogs can partition data assets by their content and anticipated audience
Data consumers typically do not know how to manipulate the data

Question 19. Which of the following does not represent a data integration pattern:

Data virtualization
Data replication
Data lineage
Message-oriented movement
Bulk/batch

Question 20. Which of the following is not a Data Movement and Integration Job Design consideration?

Design for reusability
Deployment models (e.g. containers, Kubernetes orchestration, OpenShift)
Design for parallel processing
Everything should be programmed in Python
Design for job portability (build once and run anywhere)

Question 21.Data consumers can first start to provide feedback to the current data sprint in the stakeholder review meeting.

True
False

Question 22. Which of the following could be found in catalog?

Code
Business terms
Data rules
Source data
Data lineage

Question 23. All issues need to be remediated before moving on to the next data sprint.

True
False

Question 24. Improvements to the DataOps process could involve changes to

Technology used in DataOps
DataOps team roles and responsibilities
Processes for ETL
All of the above

Question 25. Reviewing the Establish Baseline Process should include reviewing how effective are the processes for establishing a baseline for –

External Regulatory requirements
Organization maturity and Readiness
Governance and Oversight
All of the above