‘Sprawl’ is one of those overused terms in virtualization. Although to date, in my opinion, there is not a better description out there that can conjure up fear and groans in equal measure. Simply put, it is synonymous with ‘management overhead’!

Today I’m going to write about another kind of sprawl (like you really need another kind of sprawl!) - Virtualization Vendor Sprawl.

Over the years many niche players have arisen to fill in the technology gaps that were left open by mainstream virtualization. In most organizations the process of vendor consolidation is one that most growing or acquisitive enterprises will face from time to time. It’s not a task that is relished by those involved and it always proves to be a tightrope walk of cutting costs without cutting corners.

If you’ve already been through this exercise then there is not much I can add to your experience with this article. But I would ask that you share your experiences at the end of this article, for the benefit of our readership. For those about to take this route, I hope I can give you some useful points to consider.

First, what are the benefits of vendor consolidation? Vendor Consolidation has many advantages if it is done correctly, such as:

• Reduced number of contracts/documents
• Reduced number of contacts
• Easier negotiation/purchasing/procurement
• Better discount levels
• Reduced maintenance costs
• Renewals streamlined
• Service levels are consistent
• Support is consistent
• Product interaction/integration should be simplified

Sounds great doesn’t it? But before you throw out that entire extraneous product and give all your money to the largest gorilla in the room, here are a few things to consider:

• How do we get to that point?
• Are you trading-off against ‘best of breed’ technologies?
• Does vendor ‘lock-in’ give you the best deal all-round?

In addition to looking at the technologies that a company offers, in this day and age you  also need to look at their balance sheets and financial stability to make sure that they are capable of going the distance with you.

When collecting technologies together, some products may not fit neatly into categories in the same way their physical counterpart technologies did. For example, does virtualization backup fit with virtualization technologies or backup technologies, or neither? Does it fit with DR technologies?

Are you at risk of trading off ‘best of breed’ technologies? This is always the biggest concern; throwing the baby out with the bathwater so to speak. Despite virtualization being relatively new, it is mature enough to have multiple vendors with overlapping technologies. This means you should probably investigate how many candidate vendors can give you more. You probably won’t find a vendor that will deliver everything, but it makes sense to choose one that can deliver on a lot of things and is not focused on a single, specific aspect of virtualization.

This leads to the last point, vendor lock-in. If you bet the farm on one single vendor, then this can prove painful if not risky. It could affect your agility and ability to manage your virtualization investment. Just remember though, this process is designed for those who are in this for the long-haul. With that in mind the vendors you choose should be selected with that in mind and therefore start-ups will probably not figure in your list. The larger companies are capable of weathering the economic storms. But also, they can make the strategic acquisitions that ultimately will add value to your existing investment.

Vendors such as Oracle, Microsoft, VMware and Quest continually make acquisitions to help their customers “bridge the gap” by ensuring that best of breed is not lost by vendor consolidation.

Please feel free to share your experiences.

-Paul Martin

The announcement of vSphere was a great move forward for VMware.  But when you peel back the layers of vSphere it becomes evident that some of the same issues still exist for customers and most customers are going to need solutions for Monitoring, Backup, Replication, Conversions and Optimization that extend and embrace vSphere.  vSphere has expanded the number of tools that VMware offers in these areas, our solutions will continue to extend and enhance the capabilities of VMware platforms, including vSphere.  Over the next couple of months I’ll address each Vizioncore product and its fit into the VMware landscape including vSphere.  For today I’m going to focus on Monitoring.

vFoglight will work with vSphere out of the box and will be supported once we get a chance to Q/A it with the GA product from VMware. This is probably the first product I’m asked about when it comes to how vCenter will eliminate the need for vFoglight. I had the distinct pleasure to go to the vSphere labs in Cannes, FR. I will add in a side note that having proctored a lab at the 2005 VMWorld, these guys were great. I had a chance to look at the performance metrics and there were really very few changes. I tested the vCenter 2.5 U4 new tab performance chart and they were reminiscent of Lance Berc’s great tool, vmkusage. Lance is a great guy and helped me out on many customer visits. The tool sits on java and provides some basic side by side metrics, handy for basic metrics.

So why the heck would you need vFoglight? For starters, how about being able to monitor your whole VMware environment, including multiple virtual centers and have the alerts make sense. Or how about being able to drill all the way down from the Virtual Center level to see the processes within a guest to easily find an issue? vFoglight also lets you discover which of your VMs are over provisioned.  It’s a good feature to have when the business unit manager comes to you, stomps their feet and says “I NEED 4 GB OF RAM IN THAT VM!”.  You know better, but you give it to them anyways because of the 8th layer in the osi model (also referred to as the political layer).  vFoglight will alert you to that over allocation and give you raw facts helping you get approval to pull back the resources allocated to that VM. Oh, and did I mention that the intelligence in the alerts was written by Scott Herold, co-author of the Advanced Technical Design Guide for 2.5 and VI3? If you’ve read his books, the real world approach to monitoring your VMware environments should be evident.

I could go on about creating dashboards, custom reports, and monitoring virtual & physical systems, but I’m not. However, I am going to tell you that we have some exciting additions to vFoglight on the horizon. And if you need to monitor your environment today, download vFoglight and connect your Virtual Center(s) to it. We don’t load additional agents on the esx hosts, instead you install a collector on the Virtual Center server and the only other agents would be if you want to see guest processes. While you’re waiting for VMware’s advances, give vFoglight a try, I think you’ll find it a complementary solution to the power of vCenter.

Quick summary, I  want to reemphasize that even though the name and messaging changed, the challenges of working with VMware remain relatively the same.  And perhaps even stepped up a bit, but the underlying doing more with less and working smarter messaging is, to use a Talking Heads term, “same as it ever was”.   In my mind, that’s not a bad thing. VMware continues to crank out new innovations on a solid base that supports operating systems that cover the gamut of real world needs, which is just one of the reasons I  think they continue to be so successful.

Until next time…virtually speaking,

George Pradel

Virtualization and cloud computing have received a lot of attention recently. As part of the release of vSphere 4, VMware is beginning a marketing campaign centered on cloud computing. What is cloud computing and how does it apply to virtualization?

“Cloud” usually refers to the Internet, and people often depict it visually as a cloud. Cloud computing can be thought of as using different types of computing resources over the Internet. A few types of cloud computing are: Infrastructure as a Service (IaaS), Software as a Service (SaaS), and Platform as a Service (PaaS). IaaS gives access to remote computing resources. Amazon EC2 is the most prominent example of this. SaaS provides software, either client applications or web services. Those who have used Google Docs have used SaaS. PaaS allows application developers both develop for and execute code on a remote platform. Force.com is an example of PaaS.

Depending on your role, the cloud can be either abstract or concrete. The user is always outside of the cloud; they are unable to see the infrastructure behind the resource that they are using whether it is hardware, software, or storage. On the other hand, from the point of view of the engineer the infrastructure is very apparent. He or she is inside of the cloud and is helping to keep it available to the user. Where does virtualization fit in?

Virtualization is a cloud enabler, which makes it easier to provision and manage resources. For example, in enabling IaaS with virtualization users can request a machine with their desired hardware profile and usually receive it within moments. With virtualization SaaS is more flexible and reliable. Pre-configured application servers can be created to meet demand and participate in fault tolerant configurations hard to achieve without virtualization. Lastly, distributed processing nodes based on virtualization maximize scalability in PaaS. These are just a few examples of how virtualization enables cloud computing.

I’ll enjoy following the news and discussion around VMware’s cloud initiatives. I hope that this has helped to understand cloud computing and see how virtualization is an enabler for it. As our software becomes a client of cloud-based services, or helps the engineer to more easily manage cloud-enabling technology, one thing is clear: cloudy days are ahead.

Sincerely,

David Feathergill