Love the interface!
Love the data!
Love the ability to rapidly drill down to key metric or component that is the root cause of host or guest performance issues!
Love the rapid, simple appliance-based deployment!
A few observations / suggestions / questions:
Seems like it would be nice to have basic SMTP/SNMP alerting. If any object (capacity, health, workload) goes from green to yellow for instance, it would be great to generate a message or trap.
Would like to see better export in the Detail Metrics area. If I have 4 metrics selected in the View pane, it would be nice to render all of them and not just each one individually. Am I missing on how to do this?
Do permissions to the web console come from vCenter role assignments? That seems to be how things function.
Occasionally when attempting to login in to vCtrOps admin web page, once you enter the Admin account credentials you land on the vCenter Ops dashboard. Seems to happen occasionally.
Any plans for reporting capability in next release so that you can schedule recurring detail metrics reports? Similar to CapIQ reporting interface?
Overall - a very useful product!!!
I believe that the SMTP alerts there are just for administrative alerts related to the status of the vCenter Ops server. At least, that is my interpretation of the documentation and interface.
Tristan, I am happy to hear you like the product. We (the dev team) are listening to our community to hear all the things the like, dislike, and wish for in the product. Here are some responses to your questions:
You are correct that currently outbound SMTP is designed to only send out administrative alerts about vcops. There are plans to add performance and capacity alerts to the product via SMTP and SNMP.
Good suggestion on the metric export. You are correct that currently you can only export one at a time. I filed a feature request for this. 🙂
You are correct, object permissions are inherited from vcenter. Our goal was to make it as simple as possible.
Reports are also in future plans, something like the CIQ reports.
Does vCOPs require CAPiq?
How much of the Integrien product has been updared to the vCOPs or is this primarily a rebranding fo the product as a VMW offering?
What's the plan for multiple vCenter environments - is the Enterprise product??
Actually, a lot of work has gone into the development of vCenter Operations. Obviously the analytics engine is a major component but a big focus has also been on integrating the analytics with the vSphere health models that we have been working on for quite a while.
In terms of VC Ops editions, Standard and Advanced edition are targeted at small to medium size (up to 500 VM) vSphere-only environments and Enterprise edition is targeted at large, heterogeneous environments (typically multiple VCs). There is a good and very detailed comparison of what's in each edition on our website.
You can still buy vCenter CapacityIQ as an a-la-carte item but it also provides the capacity optimization capabilities in Advanced and Enterprise edition. Standard edition only has basic capacity metering.
We only have 100 VM's, but we have 2 datacenters, each with their own vCenter (currently using linked mode), ready for SRM. It's kind of s step back to have to run 2 vCOPs..
It would be great if we could monitor it with the 1 vCOPs.. c'mon.. single pane of glass and all that.. 😉
Being a smaller shop, I dont know if the boss would fund the enterprise version.
So two points about my question
1) I guess I was asking how much the analytics engine changed from when it was integrien's product, etc.
2) Thanks for the comparison link - its helpful but i think I am missing on how a large enterprise could use vCOPs with 10-20 virtual centers or a few thousand VMs (1000+)?
I like that the Enterprise is heterogenous but I am looking to see if Standard or Enterprise will scale and allow for a Stoplight/KPI type drill down into multiple vcentersand multiple data centers or clouds, etc.
Do you guys have any info on what your testing vCOPS Enterprise with - the number of virtual centers and the scale of the VM deployments (I am less interested at the moment in non-VMW hypervisors, etc).
While vCenter Operations Standard is based largely on the Integrien analytics, we spent a good deal of time integrating additional VMware logic and concepts into the product. There are many new calculated stats in the product to help correctly model workload, health, and capacity. This means for you all that you are getting some of the best of both Integrien and VMware. Of course, our work is still in progress and future versions will only get better.
While I cannot tell you specifically the target numbers of vCenter Operations Enterprise in terms of VCs and VMs, I can tell you that we have scale tested to many million stats. Depending on the structure of your VC (Host to VM ratios, number of datastores, datacenters, etc), this can translate to support for several 10,000s of vms. Contact your sales person to help understand the details behind these statements.
vCenter Oprations Enterprise currently does not have the same "Stoplight/KPI type drill down". We would be foolish not to close this gap. (Ok, so we are working on it, but I cant tell you officially when or if it will ship... )
We are aware of the issues of scale and the gaps in the product. It takes some time to merge technologies, but believe me that we are working diligently to bring everything together.
I hope this helps.
thanks for the response - we put up vCOPS Standard this week - its attractive from a UI perspective but didn't scale to your needs - we run multiple VCs and need a product that provide that view - so I look forward to the vCenter Operations Enterprise.
But it sounds like Standard vs. Enterprise is almost two different product offers? Enteprise sounds like it won't look and feel like Standard.
I am coming from the VMware Standard vs. VMware Enterprise naming system - where they do the same thing - Enterprise just does it bigger or better.
So do you know if there will be a larger version of Standard - one that can scale to multiple virtual centers and maintain that "Green/Yellow/Red" approach?
Maybe call it vCOPs "Bigger than Standard but not quite Enterprise" - minus out heterogenous management (hyper-v, xen, etc) and the collection from other frame works (BMC, CA, etc) - just a bigger version of Standard.
I figure Integrien had a product roadmap prior to the Aquisition by VMW and some of this is coming from that - I also see the stuff coming in from EMC (FastScale, Application Discovery Manager, Server Configuration Manager and Service Manager) but that's not in vCOPS Enterprise - right???
I get the DB calculations and the statistic theory (how many rows of data can fit in the database, etc) - but what is the biggest enviroment you are collecting from - can you share that??
I am happy to connect offline as well - we are a pretty large VMware environment and the management gaps/issues of scale - are really problematic for us but I recognize it as a part of the maturing of virtualization.
Yes, we are planning on supporting multiple VCs and higher scale in future releases, particularly for the Advanced version. This be able to meet your desire for a happy middle ground where you get the simplicity and ease-of-use of Standard with a higher scale and support for multi-VC.
That would be great - especially given the significant price difference between Standard and Enterprise (Comma) Standalone.
Do you know if Advanced is priced more like Standard than Enterprise (Comma) Standalone?
And how different is the analytics between the Standard and Enterprise (Comma) Standalone specific to Vmware analytics vs. non-VMW stuff?
Am I right in assume Standard and Advanced will be the same analytics??
I just want to answer the question about the analytics. The analytics engine is exactly the same between Standard, Advanced and Enterprise. In addition, the vSphere adapter (the thing that pulls data from VC and ESX) is also exactly the same. So all the stats and derived metrics and such that you see in VC Ops Std (like the workload, health and capacity badges) are also there in Enterprise.
For this specific example, the difference between Std and Ent is the UIs (but of course Enterprise has many more features than Std). Std has a UI that is customized to vSphere objects. Enterprise, since it can monitor any type of object (VM, network switch, application, etc), must have a more generic UI. However, we are planning on porting some of the Standard UI views into Enterprise so that Enterprise can display them when looking at vSphere objects.