Exam AWS Certified Solutions Architect - Professional topic 1 question 30 discussion

Answer is D Use a separate ELB for each instance type and distribute load to ELBs with Route 53 weighted round robin

only with sens

Correct answer is B

Answer is D, The company have already purchase Reversed Instance (RIs) of C3, so turn off ore remove it will waste money

D. Use a separate ELB for each instance type and distribute load to ELBs with Route 53 weighted round robin.

A - looks like the most viable option, though you need to give up RIs for c3.2xlarge instance (which is not cost effective, as you have to pay for it, unless there is an exception and Support cancel it at early stage.) It's also the most effective use of the EC2 capacity, as there is auto-scaling, meaning capacity meets the dynamic demand. Last but not least, if you check pricing calculator, c3 is way more expensive instances then m1.  B - No. Not scalable. Not cost effective to shut down RIs in havy utilization.C -  No. latency based routing is for multi region deployment where as here is just multi AZD - No. No point to buy one more ELB; weighted round robin policy is used for a different use case, e.g. associate multiple resources with a single domain name + software testing. We don't have this requirement. In addition to that, it's not the most effective use of the EC2 capacity, as there is no auto-scaling.

Answer is D Use a separate ELB for each instance type and distribute load to ELBs with Route 53 weighted round robin Considered the use of RIS all pay for use on or off instance

It's D

D Correct

I go with D as there are RIs

D is correct

D. although not very cost effective with 2 ELBs but D solves the requirement with weighted R53 routing.

A, B if you have already purchase Reversed Instance (RIs) so if you shutdown either of them, you still lose the cost. C. It does not work well, as Route53 can't route to EC2 unless it has EIP because its IP will change when rebooted. You will need EIPs for all EC2s. So it is impossible solution. D make sense, the traffic for each instance was loaded-balance however the improper type between C2 and M1 cause the CPU on M1 meanwhile no load on C2. You need another ELB, one for C2 group and one for M1 group; then route the traffic 80% to C1 and 20% to C1 for example.

I think C is the correct answer. The question asks for the "most cost effective" option. Option C removes the ELB which is cheaper than option D which adds another ELB. Additionally, Option C utilizes Latency Based routing which ensures none of the four instances are overloaded.

Answer is D

D is correct

go with D