vSphere performance and capacity management

[jeffcapes]

For some best practice research I am doing (for someone who doesnt support VMware infrastructure and software personally), at a practical level, what would be some examples that could demonstrate that a companies VMware/infrastructure engineers are taking capacity and performance management of their virtual sever infrastructure environment seriously and following best practices? 

For example, what actual evidence could be provided to demonstrate to management/internal risk departments, that your capacity/performance monitoring processes actually align with general good practice?  If this is high on your agenda and your area of responsibility, what specifically does (or should) the process involve on a daily/weekly/long term basis?

Or from another perspective, are there any red flags/weaknesses that could identify areas for improvement in this area? N/B this is primarily specific to virtual server and opposed virtual desktop infrastructure, running 100+ windows based VM. VM's run a wide range of services from file servers, SQL Server, application servers, AD, etc.

[maksym007 · Accepted solution]

Are you deploying On Prem environment or in Cloud or even Hybrid. 

When On-Prem, In this case, better to divide the applications workload into clusters. How many sites/datacenters will be there. Disaster Recovery/ Replication. 

Storage type when vSAN it can be managed by VMware admins, when SAN or NAS from my experience this Storage/Backup team.

Well-known example:

SQL- one cluster,

SAP Hana - second cluster. 

Windows + VDI can be in third and in one cluster too. 

Everything depends on the customer and costs. 

[jeffcapes]

In this case on-prem. It is more the monitoring of performance and capacity I was interested in, rather than technical design. Some places seem to have a larger daily checklist for reviewing capacity/performance related alarms and alerts triggered and then a strategic meeting to check for major issues/growth whereby capacity will be consumed without intervention within X days etc. 

[Kinnison]

Hi,

Excuse me for telling you but IMHO your question starts from a wrong approach, first of all you have to know the customer's IT context in depth, because someone will have built it according to the customer's expectations and ability (or willingness) to spend. Based on this premise, that things have been built in a way (even mostly) different from the others by no means, implicitly and demonstrably, let it be understood that the "admin/engineers" are somehow "incompetent" or "negligent".

Simply put, pointing fingers at others is mostly counterproductive, it takes nothing to build a hostile environment where the favorite sport would become to document your gaps. And then, sure that "performance and capacity" depend on "technical design", a bad project cannot produce good results, that would be an oxymoron.

Regards,
Ferdinando