Some things you'll need to consider when answering:
1. Each VNX is listed as having a maximum throughput, which is given
2. MirrorView/A is used, which means you'll need to take COFW activity into account
3. You're told to keep utilization under 70%, which will affect the maximum throughput numbers
Please attempt to answer the question, and show your calculations in detail. That way, if you're making errors somewhere, we can see where they are and help you.
Can you share what about this question you don't understand? If you show your current work or clarify your question, it will be easier for other Proven Professionals to provide an answer.
You're correct with the utilization calculations. Let's look at the rest:
The VNX at the local site processes local application I/O PLUS the writes mirrored from the remote site. Similarly, the remote VNX processes remote application I/O PLUS the writes mirrored from the local VNX.
Local VNX: 2,000 IOPs + 6 x (400 remote writes/s) = 4400 IOPs. Total capacity is 7000, therefore the headroom is 2600 IOPs.
Remote VNX: 1,200 IOPs + 6 x (500 local writes/s) = 4,200 IOPs. Total capacity is 4,900 IOPs, so the headroom is 700 IOPs.
andre_rossouw
62 Posts
0
November 3rd, 2014 07:00
Some things you'll need to consider when answering:
1. Each VNX is listed as having a maximum throughput, which is given
2. MirrorView/A is used, which means you'll need to take COFW activity into account
3. You're told to keep utilization under 70%, which will affect the maximum throughput numbers
Please attempt to answer the question, and show your calculations in detail. That way, if you're making errors somewhere, we can see where they are and help you.
KEHutchinson
633 Posts
0
October 24th, 2014 09:00
Can you share what about this question you don't understand? If you show your current work or clarify your question, it will be easier for other Proven Professionals to provide an answer.
Pwarnock1234
4 Posts
0
November 3rd, 2014 07:00
I passed this exam last week. Thanks anyway
andre_rossouw
62 Posts
0
May 18th, 2015 13:00
You're correct with the utilization calculations. Let's look at the rest:
The VNX at the local site processes local application I/O PLUS the writes mirrored from the remote site. Similarly, the remote VNX processes remote application I/O PLUS the writes mirrored from the local VNX.
Local VNX: 2,000 IOPs + 6 x (400 remote writes/s) = 4400 IOPs. Total capacity is 7000, therefore the headroom is 2600 IOPs.
Remote VNX: 1,200 IOPs + 6 x (500 local writes/s) = 4,200 IOPs. Total capacity is 4,900 IOPs, so the headroom is 700 IOPs.
Answer (A) is correct.
jony_pi
7 Posts
0
May 18th, 2015 13:00
Let me know if im right with my answer:
• At the local site, a VNX capable of performing 10,000 LUN IOPS * 0.70 = 7000 IOPS (capacity at 70% of utilization)
• At the remote site, a VNX capable of performing 7,000 LUN IOPS * 0.70 = 4900 IOPS (capacity at 70% of utilization)
Local 2,000 random 4 kB IOPS, with a R/W ratio of 3:1.
500 write iops * 6 CoFW=3000 IOPS for app to be mirrored
7000 - 3000 = 4000 IOPS available for others local apps
Remote 1,200 random 8 kB IOPS, with a R/W ratio of 2:1.
400 write iops * 6 CoFW= 2400 IOPS for app to be mirrored
4900 - 2400 = 2500 IOPS available for other remote apps