Your development teams release new versions of games running on Google Kubernetes Engine (GKE) daily. You want to create service level indicators (SLIs) to evaluate the quality of the new versions from the user's perspective. What should you do?
A.
Create CPU Utilization and Request Latency as service level indicators.
B.
Create GKE CPU Utilization and Memory Utilization as service level indicators.
C.
Create Request Latency and Error Rate as service level indicators.
D.
Create Server Uptime and Error Rate as service level indicators.
IMHO c) is the correct answer, not a).
Reason is, that we have to take the users perspective (according to the given question and also to understand what the user expects from us).
The question might be whether to choose c) or d). Considering that our requirement is to minimize the latency, we have to choose c) - that means we are striving for no downtime of the service.
yeap, definition of SLI "While many numbers can function as an SLI, we generally recommend treating the SLI as the ratio of two numbers: the number of good events divided by the total number of events" -> https://sre.google/workbook/implementing-slos/#:~:text=or%20product%20manager).-,What%20to%20Measure%3A%20Using%20SLIs,-Once%20you%20agree
I'm between C and D.
Can anyone explain why not D?
"Because it's deployed to K8s" is not an answer. I've been deploying to GKE for years, and the serve might be down just because of a missing or reconfigured property or say something is wrong with a database (e.g., exclusive table lock).
Answer is A - because Latency based SLI provides insight on user experience, while the CPU utilization SLI provides insight for the operations team in case the latency is high.
CPU utilization doesn't tell us about the user experience except perhaps if it hits 100%. The errors & latency (driven party by CPU ) are much better indicators
The correct answer is: C. Create Request Latency and Error Rate as service level indicators.
Request Latency measures the time it takes for a request to be processed by the game, and is an important indicator of the responsiveness of the game from the user's perspective. A high request latency can indicate that the game is experiencing performance issues or is under heavy load, which may negatively impact the user experience.
Error Rate measures the percentage of requests that result in errors, such as HTTP errors or timeouts. A high error rate can indicate that the game is experiencing technical issues or is not able to handle incoming requests effectively, which can also negatively impact the user experience. By monitoring these SLIs, you can identify issues with the game's performance and take appropriate action to improve the user experience.
You can use tools such as Stackdriver Monitoring or Cloud Monitoring to create custom metrics and alerts based on these SLIs, and track the performance of your games over time. You can also use these tools to set up automated responses to alerts, such as automatically scaling the number of game instances up or down based on the current load.
A voting comment increases the vote count for the chosen answer by one.
Upvoting a comment with a selected answer will also increase the vote count towards that answer by one.
So if you see a comment that you already agree with, you can upvote it instead of posting a new comment.
XDevX
Highly Voted 3 years, 4 months agomedeis_jar
3 years, 1 month agoMamthaSJ
Highly Voted 3 years, 4 months agoVinaykas
Most Recent 3 months, 1 week agoGino17m
6 months, 4 weeks agodon_v
10 months, 1 week agomustafa1p
10 months, 3 weeks agothewalker
1 year agoVignesh_Krishnamurthi
1 year, 4 months agoBiddlyBdoyng
1 year, 5 months agoalekonko
1 year, 7 months agoDeb2293
1 year, 8 months agosomchaikin
1 year, 8 months agoomermahgoub
1 year, 11 months agoomermahgoub
1 year, 11 months agosurajkrishnamurthy
1 year, 11 months agomegumin
2 years agoAzureDP900
2 years, 1 month agochickennuggets
2 years, 3 months ago