v9 Trial install - Understanding reports and stats

[Blue407]

Hi All

I'm currently testing Veeam One and Backup & Replication.

I've left Veeam One running for a couple of days and am now going through the alerts and reports, to see what it's noticed.

1. One my primary Hyper-V Server (HS101, running Server 2012) its reporting high Memory Pages Rate, Avg: 5547.20, Max: 57910.00 and Recommended < 500.
This has occurred multiple times and the recommendation is to add more memory to the Host.
This host has 192GB of RAM and the report itself shows that available memory is: Avg: 20.42%, Max: 51.06% and Recommended < 80%. So what's the problem?

2. Same server, Host Memory Pressure shows: Avg:65.07, Max: 165.00 and Recommended < 80. Any suggestions?

3. Another Host (DUF, Server 2012 R2) also shows high Memory Pages Rate, Avg: 149.71, Max: 1956.00 and Recommended < 500. This server is not yet in production, I'm just testing with it first. I has 2 VM's and they are not doing much at all. The most taxing thing will be Hyper-V Replication to another local server every 5 mins. However almost no data is changing, so little replication required.

4. Same host (DUF) again, this time high disk Latency, Read for C: drive is , Avg: 3.00, Max: 19.00 and Recommended < 15. And Disk Write Latency for C:, Avg: 25.67, Max: 82.00 and Recommended < 15. And Disk Write Latency for D:, Avg: 23.27, Max: 58.00 and Recommended < 15.

Any suggestions of ideas on all the above would really be appreciated. I need to understand if there is something I can tweak or improve.

I also find the reports do not format well on screen, often missing out rows at the bottom of the page, when I printed them, they scrolled onto 2nd page...
Is there any way to get the report to auto-login, it's a pain having to enter permissions every single time I run one?

[Shestakov]

Hi Paul and welcome to the community!
1) You may have some under-provisioned VMs which cause use of memory pages. Try to use a corresponding report to check if you have such VMs.
2) Same recommendation. I would also check what time you had those alarms triggered and if it`s related to backup activity.
3,4) Do you use off-host or on-host proxy for the replication? Since average value is not that high you may have some pikes during backup jobs.
There is a related article with more information. Please take a look.
Thanks!

[Blue407]

Hi Paul and welcome to the community!
1) You may have some under-provisioned VMs which cause use of memory pages. Try to use a corresponding report to check if you have such VMs.
2) Same recommendation. I would also check what time you had those alarms triggered and if it`s related to backup activity.
3,4) Do you use off-host or on-host proxy for the replication? Since average value is not that high you may have some pikes during backup jobs.
There is a related article with more information. Please take a look.
Thanks!

Thanks for the warm welcome

I have run the Undersized VM Report for the whole Virtual Infrastructure, out of 42 VM's none are undersized.

3,4 - the Replication is using native MS Hyper-V Replication tools, not Veeam. Its worrying that this server is showing such poor performance and its not even in production yet...
In case it is relevant, its a Dell Poweredge 320 with 4 x 2TB 7.2k SAS drives running RAID 5. I would have thought with 4 spindles, performance would have been adequate.

I will read through the link and see if that helps

[Shestakov]

1-2) What about the timing of pikes? Does it related to backup jobs?
3-4) I would also run Configuration Assessment report to check if the infrastructure is tuned up to the best practices.
Thanks!