Which three parameters should be considered when calculating the bandwidth for vSphere Replication? (Choose three.)
A.
Data change rate
B.
Traffic rates
C.
Link speed
D.
Application type
E.
Hardware type
Explanation:
Explanation/Reference:
A, B, and C are correct:
http://pubs.vmware.com/vsphere-replication-61/index.jsp#com.vmware.vsphere.replication-admin.doc/GUID-4A34D0C9-8CC1-46C4-96FF-3BF7583D3C4F.html?resultof=%2522%256c%2569%256e%256b%2520%2573%2570%2565%2565%2564%2522%2520
1
0
6.0 version
https://pubs.vmware.com/vsphere-replication-60/index.jsp?topic=%2Fcom.vmware.vsphere.replication-admin.doc%2FGUID-4A34D0C9-8CC1-46C4-96FF-3BF7583D3C4F.html&resultof=%22bandwidth%22%20
0
0
Although VMware documentation mentions traffic rates (answer ‘B’ above) as a factor, technically it’s not a parameter to be accounted for ***to calculate*** bandwidth requirements. Traffic rates belong to the cost-related aspect.
Also, the documentation states: “To determine the bandwidth that vSphere Replication requires to replicate virtual machines efficiently, you calculate the average data change rate within an RPO period divided by the link speed.” – WRONG !! – because such calculations produce how many links (or fraction of the link) you need in order to transfer data within RPO period.
In real life you would take average-changed data withing RPO and divide it by the RPO window to calculate required bandwidth. E.g.: for 10 GB of changed data and 1 hour RPO, 10 x 10 (bytes to bits) x 1000 (Gb to Mb) / 3600 sec = 28 Mbps of dedicated bandwidth is required. Of course, this is a bare minimum that should be at least doubled to ensure replication completes before the next sync.
0
0