C_TSCM44_65 Exam Discussions & Posts
MAGICBUNNIEZ
Commented on July 26, 2025
194: To ensure the organization follows security best practices when transitioning an on-premises application to a cloud-based service, the security team should implement all of the listed controls, as they address different critical aspects of cloud security. However, if prioritizing the most fundamental requirement for a secure cloud deployment, the team should first ensure:
1. Virtualization and Isolation of Resources
Why? Cloud environments rely on multi-tenancy, so virtualization and isolation prevent cross-tenant attacks (e.g., VM escapes, side-channel attacks).
Best Practice: Use hypervisor security, container isolation, and micro-segmentation to enforce strict boundaries between workloads.
Supporting Controls (Also Critical but Secondary to Core Isolation):
2. Network Segmentation
Limits lateral movement if a breach occurs (e.g., separating frontend/backend tiers).
However, isolation at the virtualization layer is more foundational in cloud environments.
3. Data Encryption
UNITED STATES
Josh
Commented on July 26, 2025
for Q483 The company needs to run a Windows containerized .NET 6 job every 10 minutes, with a runtime varying between 1 and 3 minutes, in the AWS Cloud, and do so MOST cost-effectively.
Let's evaluate each solution:
Create an AWS Lambda function based on the container image of the job. Configure Amazon EventBridge to invoke the function every 10 minutes.
AWS Lambda for containers: Lambda now supports deploying functions as container images, including Windows containers. This is a serverless option, meaning you don't manage any servers.
Runtime limits: AWS Lambda functions have a maximum runtime of 15 minutes. Since the job runs between 1 and 3 minutes, it fits well within this limit.
Cost-effectiveness: Lambda's pricing model is based on duration and memory consumed, billed in millisecond increments. You only pay when your code is running. For a job running only 1-3 minutes every 10 minutes, this is incredibly cost-effective because you pay almost nothing for idle time. EventBridg
UNITED STATES
Ash
Commented on July 26, 2025
great studying from portal.
CANADA
q
Commented on July 26, 2025
helpful question
UNITED STATES
stephen
Commented on July 26, 2025
very helpful
Anonymous
u
Commented on July 26, 2025
helped in prep
UNITED STATES
Jay
Commented on July 26, 2025
Looking good so far
Anonymous
rey
Commented on July 26, 2025
reliable questionaires
EUROPEAN UNION
rey
Commented on July 26, 2025
very interesting topics, this helped me a lot
EUROPEAN UNION
chatbot
Commented on July 26, 2025
The questions offer a great help in preparation. These are related to topics in exam
Anonymous
prasanna
Commented on July 26, 2025
The questions are very helpful after completing the study materials and before the exam! Thanks
UNITED STATES
rey
Commented on July 26, 2025
usable contents, very helpful
EUROPEAN UNION
rey
Commented on July 26, 2025
great questions, reliable
EUROPEAN UNION
Suchi
Commented on July 26, 2025
great content
Anonymous
Suchi
Commented on July 26, 2025
Great material
Anonymous
mark
Commented on July 26, 2025
amazing contents
EUROPEAN UNION
gm
Commented on July 26, 2025
Question 123 - False -
Reader accounts in Snowflake are designed for accessing and querying data shared by a provider, but they cannot extract that data for use outside of the Snowflake environment. They are primarily used for data consumption within Snowflake.
Anonymous
Atul
Commented on July 26, 2025
Answer to Question 55 is A GAN as it is used for the data generation. Option C is not correct.
Anonymous
gm
Commented on July 26, 2025
Q109 - False because owned by role not by user
Anonymous
Alex
Commented on July 26, 2025
very good questions
ROMANIA
gm
Commented on July 26, 2025
91 - Answer should be True
AWS PrivateLink can be used to establish a secure, private connection between a customer's on-premise data center (via AWS Direct Connect) and Snowflake. This connection bypasses the public internet, enhancing security and reducing the risk of data exposure.
Anonymous
Amit
Commented on July 26, 2025
Detailed and specific - stick to the foundational themes to test your knowledge and you will succeed
Anonymous
mark
Commented on July 26, 2025
very informative
EUROPEAN UNION
gm
Commented on July 26, 2025
Question 77: Answer should be all
Bulk copy from an External Stage:
This involves using the COPY INTO command to load data from files stored in external cloud storage locations (e.g., Amazon S3, Azure Blob Storage, Google Cloud Storage) into a Snowflake table.
Continuous load using Snowpipe REST API:
Snowpipe enables continuous data loading from staged files as soon as they become available. The Snowpipe REST API can be used to trigger the loading process for new files in an external stage.
The Snowflake Web Interface (UI) data loading wizard:
Snowsight, Snowflake's web interface, provides a user-friendly wizard that guides users through the process of loading data into tables, often from local files or staged files.
Bulk copy from an Internal Stage:
Similar to external stages, data can be bulk loaded from internal stages (storage locations within Snowflake) into tables using the COPY INTO command. Files can be placed in internal stages using commands like PUT.
Anonymous
gm
Commented on July 26, 2025
Question 68:
Explanation:
INSERT:
When inserting data into a table, Snowflake usually writes to new partitions, allowing other queries to access the existing data simultaneously.
COPY:
Similar to INSERT, the COPY command often writes to new partitions, enabling parallel execution with other operations.
Why other options are incorrect:
UPDATE:
This command modifies existing data in a table, which requires locking the affected rows to prevent conflicts. This makes it a blocking operation, meaning other queries that need to access the same rows will have to wait until the UPDATE finishes.
MERGE:
Like UPDATE, the MERGE command modifies data in a table, requiring locks and thus being a blocking operation. It combines data from multiple sources into a target table, which can involve modifying existing rows.
Anonymous
mark
Commented on July 26, 2025
very interesting
EUROPEAN UNION
Kannamani Ramasamy
Commented on July 26, 2025
Good practice questions
Anonymous
lp
Commented on July 25, 2025
good set of questions to help prep
UNITED STATES
gm
Commented on July 25, 2025
Question 59 - False. In Snowflake, when a role is created, the role that is active at the time of creation is assigned ownership, not the individual user. This ownership can be transferred to another role using the GRANT OWNERSHIP command.
SINGAPORE
gm
Commented on July 25, 2025
Question 48 -
? They roll up and release approximately monthly, but customers can request early release application.
- While behavior changes are bundled monthly, releases themselves are weekly. Early access is available for Enterprise Edition and higher, but this applies to weekly releases, not monthly rollups.
? A customer is assigned a 30-minute window during which the system will be unavailable and upgraded.
- Snowflake upgrades are seamless and do not involve downtime. Customers are not assigned maintenance windows.
SINGAPORE
gm
Commented on July 25, 2025
Question 42 - Explaining why not others
? Snowflake tables are the physical instantiation of data: This is misleading because Snowflake abstracts physical storage from users.
? Clustering keys are required for optimal performance: Clustering keys are optional and used for very large tables to improve query performance, but not required.
? Snowflake tables are owned by a user: Tables are owned by roles, not individual users, as part of Snowflake's role-based access control (RBAC) model.
SINGAPORE
MK
Commented on July 25, 2025
excellent website for preparing for the ms 900 exam
UNITED STATES
gm
Commented on July 25, 2025
Question 37: Explaination -
Here are the three correct statements about Snowflake data loading:
? VARIANT "null" values are not the same as SQL NULL values
In Snowflake, the string "null" stored in a VARIANT column is treated as a string, not as a SQL NULL. This distinction is important when querying or transforming semi-structured data like JSON.
? It is recommended to validate the data before loading into the Snowflake target table
Validating data helps ensure integrity, prevents errors during loading, and avoids polluting production tables with bad or malformed data.
? It is recommended to use staging tables to manage MERGE statements
Staging tables allow for cleaner, more controlled data transformations and help manage complex MERGE operations efficiently.
? It is recommended to do frequent, single row DMLs
This is not recommended. Snowflake is optimized for bulk operations. Frequent single-row DMLs (INSERT, UPDATE, DELETE) can be inefficient and costly.
SINGAPORE
gm
Commented on July 25, 2025
True or False: A third-party tool that supports standard JDBC or ODBC but has no Snowflake-specific driver will be unable to connect to Snowflake.
Answer should be False because Snowflake provides standard JDBC and ODBC drivers that allow connections from tools supporting these standards
SINGAPORE
gm
Commented on July 25, 2025
Helpful questions
SINGAPORE
Ankitha
Commented on July 25, 2025
Its very useful
UNITED STATES
ADB
Commented on July 25, 2025
very relevant questions providing accurate answers which helps in certification preparation
INDIA
Ricardo Nelumba
Commented on July 25, 2025
On question 49, answer B makes more sense because Athena requires files to be in queryable formats (e.g., CSV/JSON), not audio/text transcripts. Glacier Select has high latency (hours).
Anonymous
Nayeh
Commented on July 25, 2025
Anyone passed exam recently? shade a light on questions asked from here
Anonymous
ADB
Commented on July 25, 2025
this questions provides contextual insights for exam preparations
INDIA
cc
Commented on July 25, 2025
nice questions
JAPAN
cuuns
Commented on July 25, 2025
good questioins
JAPAN
harry
Commented on July 25, 2025
great content
JAPAN
BV
Commented on July 25, 2025
The answer to question 34 is incorrect. It should be:
ALTER TABLE ...
ADD CONSTRAINT ... PRIMARY KEY NONCLUSTERED
NOT ENFORCED
Fabric does not support clustered index primary keys and does not support enforcing constraints.
CANADA
Maarten
Commented on July 25, 2025
Looking good
NETHERLANDS
Riya
Commented on July 25, 2025
Very helpful
Anonymous
Riya
Commented on July 25, 2025
Very good and helpful
Anonymous
ghostdog248
Commented on July 25, 2025
Wish you all the best
Anonymous
AL
Commented on July 25, 2025
Good content
SPAIN
Abhijeet
Commented on July 25, 2025
Good questions
Anonymous