...but not necessarily in that order. 
In this post, I hope to do a full review of my personal experience with VSC1 and 2, as well as some hopes and projections as to where I WOULD HOPE TO see it go in the future.
I wanted to preface this post with the disclosure that I am a long-time NetApp user/customer. I am very experienced with the hardware, core systems, management platforms, and bolt-on softwate to manage and report on storage-based backup & recovery of tier 1 applications.
As much as NetApp likes to boast that they are a single platform across all of their hardware, well, for that part they are right. DataONTAP is a wonderful operational platform with loads of customizable features, an easy-to-use UI in FilerView, and a powerful CLI and now API-set that you can do just about anything with.
What they DON'T tell you is that for every little thing you want to do in addition to the base management of systems requires a separate software installation. I'll list a few of them here:
SnapManager Suite (I personally call it the suite, because there's a different one for the following..)
SM-SQL - Microsoft SQL Server
SME - Microsoft Exchange Server
SMO - Oracle on Linux
SMO - Oracle on Windows
SMVI - VMware
SMHV - Hyper-V
Other's include:
Operations Manager
Protection Manager
Provisioning Manager
System Manager
In all fairness, since v3.8.x, we've seen serious improvement in the integration of OpsMgr, ProtMgr, and ProvMgr. I'm just not sure I understand the need to have 2 or 3 different products. If someone bold-faced told me it was to milk more money out of the customers, I'd completely understand, but the pricing models don't reflect that.
I'm sure I'm missing some. Anyway, each of these applications was designed by a completely different team, obviously none of which communicated with each other. Function, Reporting, Scheduling, even basic UI's are completely different in all of them. For a company who spends so much time preaching about single interface for everything, you'd think they would carry the same motto across to these add-on applications.
Now, let's not get our panties in a wad here. I'm not trying to harp on NetApp (too much). They know these areas need improving, and there is talk of something called "Unified View" coming end of next year to consolidate all of these apps into a single pane of glass.
Enter VSC...
I remember being so excited when I first heard about this coming out last year. FINALLY someone's getting the integration ball rolling. Some vocabulary for those unfamiliar. VSC stands for Virtual Storage Console, and is NetApp's integration/plugin that becomes an additional tab in your vCenter.
I remember loading it up and being so excited seeing that I could view all of my storage from within VirtualCenter. COOL! I could auto-set NFS parameters to make my hosts comply to VMware + NetApp Best Practices. COOL!
But wait, is that it? I mean it's cool and all to see all of this stuff inside of my SPOG vCenter, but ....meh....it's v1.0, I'm sure it'll get better with the next release. This is just laying the framework.
Over the course of the past year, I'm sure you've all heard more about the Rapid Cloning Utility and Snapmanager for Virtual Infrastructure products than the VSC.
I heard 2.0 was coming, was super-excited about it (again), but never got a chance to play with it until this week when I was upgrading my entire Infrastructure to 4.1. I always thought, "I'll just reload it when I rebuild VC."
From my initial views, it's not much different, to be honest. It has all the same views as it did before, about the same functionality (minimal) as before, and really the only difference it has is that they've baked in RCU and SMVI into the VSC as one thing. While that's cool and all, a lot of the core functionality of SMVI was not updated/fixed, reporting was not enhanced in any way (BIG, BIG problem), and I still do not regret my decision of moving away from SMVI to vRanger. Now, I have to eat up TERABYTES of space to store vRanger repo's instead of just having little snapshots. All because you won't change your reporting format, or the way you send out the notifications. It's not so much about that, but that the report lives on the server and dies with the snapshot. So if you expire a snapshot, it DELETES THE HISTORICAL REPORT! NO! I could (and have) rant about this for hours, but it is not the intent of this post. I'll leave it as saying, being in the Public Healthcare space, the "reporting" in SMVI is simply unacceptable, and no matter how simple and great you make the functionality of the SMVI product, we're not going to use it until that part is fixed.
You're also not going to find a lot of customers using RCU outside of the bigger VDI shops, or places that have hundreds and thousands of VM's.
My hopes for this product are vast. I really hope they see how HUGE this could be for virtual environments running NetApp, and what a huge player/selling point this could be. Because it's not at the moment.
So, let's get started.
I want to be able to customize the view somewhat by adding additional columns. Add a column for some of the pertinent stuff such as SIS (dedupe) being enabled on the volume (read: DATASTORE). Show me a column with a percentage of space I've saved with dedupe. Throw it in my face! Don't make me go and hunt for it on the command line with a 'df -s'. Autogrow and Snapreserve percentages, as well as snap space consumed would be awesome here! Stop making me go to System Manager or FilerView to set settings on a Datastore volume. Build that functionality in here. Right click on a Datastore and enable dedupe, set the schedule for dedupe, set Autogrow and Autodelete functionality for snapshots! All within vCenter! Amazing, right?! I know.
The views in each section should be different based on what tier you have selected in vCenter. vCenter, Cluster, Host, Resource Pool, VM. Each view should cater to that particular level, so that when I click a host, I only see the datastores mounted to it. When I click a cluster, I see all datastores mounted to ALL hosts in the cluster. Put a little more customization into it so that we can drill down like that.
If you select a VM, it removes the Overview section, and takes you straight into the SMVI interface. WHY?!?! You could show me SO much here about my individual VM's, such as how much that VM is consuming within my Datastore. Which brings me to my next point:
ALIGNMENT! At the VM view, you could build in 'mbrscan' (instead of making us download it) to tell me whether my VM is aligned or not, and even include the functionality to power down the VM, align it, and boot it back up! As important as alignment has become to storage performance, I cannot believe you're still making us use mbrscan and mbralign manually! How are we supposed to continue to do that with ESXi going forward?! NetApp's answer: "You can run it from a Linux VM." I DONT WANT TO!:)* Other 3rd parties are capitalizing on our follies as admins BIG TIME by SELLING us this sort of product.
On a positive note, the ability to set all of the best practices parameters is AMAZING! This is the essence of this product, and you should see this as a stepping stone on how to do other things that we would typically do in FilerView and/or at the CLI.
Another positive is that the installation and setup are super-simple! Install it on your vCenter server, register the plugin, and when you open the tab the first time, it autodiscovers all of your storage systems and prompts you for credentials. DONE! Brilliant!
You have the potential to make this a REPLACEMENT for Filerview and System Manager for those customers with Virtual Environments.
My hope is that you'll start listening to us, instead of telling me: "Sorry, we're not going to change that," as I was told in a communities post.
We're out here! We want this to be an amazing product! Let us help!
-Nick
