Capacity and Performance Trending reports

Hi, over the past couple of weeks I have been reviewing numerous posts on pulling data via powercli.  I have a customer who wants some executive monthly capacity trending reports for the past year and their engineers want monthly performance trending reports.  For the most part LucD has done a great job at compiling scripts to pull 75% of the information I need to create the excel reports.  If there is a way to simplify extracting the data would be very helpful.

It would be great if all this information could be pulled historically from vCenter but exporting it daily would be okay also.

The current environment is vSphere 4.x with plans to upgrade to 5.x in the coming months.  There is a mixture of VMFS and NFS.

This is mostly to help predict when to buy additional hardware based of past trends.

Note:  Its about time they added the CSV Append feature in Powershell v3.

What I need for the monthly capacity trending report:

• At the Cluster layer over the past year or start collecting daily.
  • Total physical CPUs, CPU cores, provisioned vCPUs to VMs and over-commit ratio. CPU count not in Mhz.
  • Total physical RAM, provisioned vRAM to VMs and over-commit ratio.  In MB or GB.
  • Total physical Disk, used Disk and provisioned disk to VMs.

Performance reports:  These I have except for the VM disk performance on NFS datastores.

• At the host layer over the past month @ 2 hour intervals.
  • Max CPU, Avg CPU
  • Max MEM, Avg MEM

• At the VM layer over the past month @ 2 hour intervals.  I have everything for this report except for vmdk disk statistics on NFS.
  • CPU max and avg
  • MEM max, avg, and ballooning
  • Disk max and avg latency

Storage:

The historical datastore capacity script by LucD has is great but for some reason a few of my datastores return null even though the data shows up in vCenter.  This could be because the environment is 4.x.

What I've started with for daily capacity collections but would like to pull this historically if possible.  Some of this doesn't pull data so it is marked "NA"

Host and VM performance history script:  Need disk counters for VMs on NFS.  CPU ready time maybe a good one to add here also.

$esxImpl = Get-VM

$todayMidnight = (Get-Date -Hour 0 -Minute 0 -Second 0).AddMinutes(-1)

$workingDays = "Monday","Tuesday","Wednesday","Thursday","Friday"

$dayStart = New-Object DateTime(1,1,1,10,00,0)     # 10:00 AM

$dayEnd = New-Object DateTime(1,1,1,16,00,0)      # 04:00 PM

$stats = Get-Stat -Entity $esxImpl -Stat cpu.usage.average,mem.usage.average -Start $todayMidnight.AddDays(-31) -Finish $todayMidnight.AddDays(-1) -IntervalSecs 7200

$report = $stats # | Where-Object {

  $workingDays -contains $_.Timestamp.DayOfWeek -and

  $_.Timestamp.TimeOfDay -gt $dayStart.TimeOfDay -and

  $_.Timestamp.TimeOfDay -lt $dayEnd.TimeOfDay

# }

$report | Export-Csv "D:\Client report\Performance\PerfData\ag-VM-performance-avg.csv" -NoTypeInformation -UseCulture

$esxImpl = Get-VMHost

$todayMidnight = (Get-Date -Hour 0 -Minute 0 -Second 0).AddMinutes(-1)

$workingDays = "Monday","Tuesday","Wednesday","Thursday","Friday"

$dayStart = New-Object DateTime(1,1,1,10,00,0)     # 10:00 AM

$dayEnd = New-Object DateTime(1,1,1,16,00,0)      # 04:00 PM

$stats = Get-Stat -Entity $esxImpl -Stat cpu.usage.average,mem.usage.average -Start $todayMidnight.AddDays(-31) -Finish $todayMidnight.AddDays(-1) -IntervalSecs 7200

$report = $stats # | Where-Object {

  $workingDays -contains $_.Timestamp.DayOfWeek -and

  $_.Timestamp.TimeOfDay -gt $dayStart.TimeOfDay -and

  $_.Timestamp.TimeOfDay -lt $dayEnd.TimeOfDay

# }

$report | Export-Csv "D:\Client report\Performance\PerfData\ag-Host-performance-avg.csv" -NoTypeInformation -UseCulture

Message was edited by: JayLB