TFD9 Pre-work: Veeam

Veeam, like Nutanix, are Tech Field Day veterans, so I expect a smooth presentation.

I’ve heard a bit about Veeam for some years now, though not in great detail. What I heard was generally glowing praise for its capabilities in backing up VMware environments.

Unencumbered by issues of backwards-compatibility, or enterprise-wide capabilities, Veeam had sprung up as the way (according to those I spoke to) of backing up virtual environments. Their product was designed with virtual environments in mind, so this makes a lot of sense. Unlike a, say, NetBackup or Legato, who grew out of Unix oriented tape environments, Veeam could ignore a lot of the complexities and start fresh, learn from their mistakes, and take advantage of newer ways of doing things.

Such is the way with the rapid pace of change in technology.

Veeam appears to be a software based addition to the hypervisor that handles block-level replication of VM image data to some sort of secondary media. You can restore whole images or files inside the VM either back to the original location, or to a different location.

None of that is really that amazing, since that’s how every backup product I’ve used since 1995 has worked. However, the Windows world lagged mainframe and Unix, so a lot of what is old hat to me was a new problem for the NT 3.5.1 folks. The rapid rise of VMware certainly added a bunch of new issues very quickly.

Restore of files from within a VM, for example, was a huge challenge for the legacy backup software providers simply because VMs were a new thing and they had to write software to handle this new use case, without breaking their existing software. With huge and complex (and often not that well written in the first place) software, regression is a big deal, so Veeam and their ilk have a natural advantage: they can change and adapt much quicker.

So for someone with a lot of virtual stuff, Veeam makes a lot of sense.

Enterprise is Legacy

I’m going to go off on a tangent for a little bit here, because I’ve had difficulties with several large international vendors (not Veeam) not understanding what enterprise companies need.

Veeam, and products like it, are probably excellent at what they do, which is to provide great focus for a specific niche. They are, by virtue of their focus, able to better serve that niche than a more general product.

But here’s the thing: in an enterprise of moderate size that’s been around for more than say, five years, you will probably not have a homogeneous technology environment. There will be VMs, yes, but they are likely to be farms running different versions. There will be physicals. Again, probably multiple different versions.

It is rare that an organisation of any size will have single-sourced all, or even a majority, of their gear from a single vendor or on a single version. As soon as you have any kind of distributed decision making on purchases, business vs. IT, or even just different bits of IT, then different people will have made different decisions about what is right for them.

Backing that up requires one of two things: a single product that can deal with all the weirdness, or lots of different products that handle each situation. If you have Veeam for your virtual environments, what do you use for your physicals? And is it a different product for Windows physicals compared to Unix physicals? What about the SANs? If you have some EMC and some Compellant, do you use different tools?

What criteria do you use to make these decisions?

There’s no one right answer, and, in my view, the superior answer has a lot more to do with the way a company is run than anything to do with technology. If the company prefers a single point of control by a low-skill operations team, then a generalist tool with few features that’s easy-to-drive is probably a good choice. However, if you have a lot of mostly independent teams full of highly skilled technical people, all with specialist needs, then several different tools, each tailored to the different teams’ use-cases is probably better.

There’s no such thing as a objectively better product. The issue at hand is whether or not a specific product solves the problem that a customer has, and that problem is almost never a technical one.

But any given product simply cannot meet both of these needs. It cannot be both a generalist tool, and a highly customised specialist niche tool. Attempts to do it always, always end in failure.

Target Markets

I feel like I’ve been banging on about target markets a bit in the past few posts. It’s a really important concept to understand as a customer, and not just as a marketer or a pre-sales engineer.

If your target market is large, heterogeneous enterprise customers, your product needs to do different things, and will be priced differently, than one aimed at mid-sized, homogeneous VMware and Windows only shops. That doesn’t make either product ‘better’. They are simply different.

Understanding how you, as a customer, are different from other customers can help you to look for the products that best suit you. Similarly, vendors understanding their own company’s target market can help them determine if this customer matches that target market.

If there’s a match, you’re likely to get both a sale and a happy customer whose needs are fulfilled and who doesn’t feel ripped off.

If there isn’t a match, getting a sale is likely to be the beginning of a painful process of disappointment on both sides.

Trying to match more than one target market (called straddling in the lingo) is always a bad idea. You end up not really satisfying either market, and people will only buy your product if there’s nothing else better out there. Targeting only one of those markets is cheaper and easier for someone else to do, so it’s (relatively) easy for them to take any of the customers who might have bought your product in the first place.

Back to Veeam

I’ll be looking out for who Veeam’s target market is, and trying to understand how they’ve developed their product to match.

I want to see evidence of how Veeam have determined what those customers want (and hopefully not gone for a Field of Dreams/”build it and they will come” approach) and how their products meet those needs.

Particularly of interest to me will be who they see themselves pitching at. Is it virtualisation admins? Backup administrators? Heads of IT? Business heads? CIOs, and of what size company?

Each of these people have different roles, and will want different things.

And lastly, is their target market changing? Where does it look like Veeam are heading in the next 2-5 years?

Should be fascinating.