You manage a cloud service that hosts a customer-facing application. The application allows
users to upload images and create collages. The cloud service is running in two medium
instances and utilizes Azure Queue storage for image processing. The storage account is
configured to be locally redundant.
The sales department plans to send a newsletter to potential clients. As a result, you expect
a significant increase in global traffic.
You need to recommend a solution that meets the following requirements:
✑ Configure the cloud service to ensure the application is responsive to the traffic increase.
✑ Minimize hosting and administration costs.
What are two possible ways to achieve this goal? Each correct answer presents a complete
solution.
A.
Configure the cloud service to run in two Large instances.
B.
Configure the cloud service to auto-scale to three instances when processor utilization is
above 80%.
C.
Configure the storage account to be geo-redundant
D.
Deploy a new cloud service in a separate data center. Use Azure Traffic Manager to load
balance traffic between the cloud services.
E.
Configure the cloud service to auto-scale when the queue exceeds 1000 entries per
machine.
http://msdn.microsoft.com/en-us/library/hh680945(v=pandp.50).aspx
Use Traffic Manager would certainly solve the issue. Why not using it? Costs maybe?
traffic manager is not for scaling but only to distribute the charge equally (round robin)
http://azure.microsoft.com/en-us/documentation/articles/cloud-services-how-to-scale/#autoscale
it it correct?
is the above answer is correct?
Joni, just leave azure alone 🙂
+1
+1
+100
poor joni
I think its correct
Yes papa!
Above answers are correct
Correct – this is a simple ‘how do you scale’ question. CPU + QUEUE Messages
I don’t think it is a simple ‘how so you scale’ question. The requirements are application responsiveness and manage hosting and admin costs.
I think it is B&D
If there is a significant increase in global traffic and you need to ensure the application is responsive to the traffic increase, you would use traffic manager to distribute the load and direct traffic to the closest data center to where the request came from.
From –
https://azure.microsoft.com/en-us/documentation/articles/traffic-manager-overview/
Improve responsiveness for high performing applications – Azure allows you to run cloud services or websites in datacenters located around the world. Traffic Manager can improve the responsiveness of your applications and content delivery times by directing end-users to the endpoint with the lowest network latency from the client.
Plus E using 1000 entries per machine seem arbitrary. How do we know that is the threshold to use?
You know what. After reading it again and considering the requirement of Minimize hosting and administration costs.
The least administration is B & E by using autoscale and deploying the cloud service to another data center is not minimizing hosting so it could not be D.