There are reports in Bloomberg and the Wall Street Journal Online that BMC is in talks with Bain Capital and Golden Gate Capital to be taken private. This is after Quest Software tried to go private last year only to end up being acquired by Dell who is now itself in talks to go private and while there are also rumors that private equity investors are looking at Compuware as well. So what is going on here?
The History of the Management Software Business
Ever since the introduction of the PC the IT industry has been swept by waves of innovation. Those waves have included the PC, the LAN (Ethernet and Token Ring), the Network Operating System (Novel, and then Microsoft LAN Manager), the first generation of Windows (3.x), the first 32 bit versions of Windows (W95 and W98), Windows NT, client/server systems and applications, continuous attempts to improve developer productivity (Visual Basic, Java, .NET, and now PHP, Python, Ruby, Node-JS and who knows what else) and most lately data center virtualization.
Each of those disruptive innovations were in general deployed on a wide scale basis across enterprises worldwide before people woke up and realized that the new thing required a new method to manage the new thing. So new companies were formed to manage the new thing. The good new companies were eventually acquired by the traditional large systems management companies (IBM, BMC, CA, HP, Quest, and Compuware). IBM acquired Tivoli which was started to manage client/server based systems. HP bought Mercury Interactive to become a top tier management software vendor. CA bought Wily to manage Java based applications. The list goes on and on.
The Seismic Shift in the Management Software Business
However, we are today in an unprecedented place in the IT industry. The following forces are reshaping the industry in unprecedented ways, and at an unprecedented pace:
- Data center virtualization, popularized by VMware has in its first generation abstracted the management and operation of CPU and memory resources from the hardware that provides these resources. This was done via the VMware hypervisor (ESXi) and the product that contains it (vSphere). This alone has resulted in the creation of an entirely new ecosystem of management vendors including VMware itself (with vCenter Operations and vCAC), Veeam, AFORE, Astute Networks, Bitdefender, Cirba, CloudBolt Software, Confio Software, Embotics, FluidOps, Greenbytes, Highcloud Security, Hotlink, Intigua, Puppet Labs, Quantum, Reflex Systems, Splunk, Teradici, Tintri, VMTurbo, Virtual Bridges, Virtual Instruments, Virtustream, Xangati, Xceedium, Zenoss, and Zerto.
- Following upon VMware’s success in pioneering the data center virtualization business, Microsoft caught up with VMware on the CPU and memory virtualization front with Hyper-V, making the management of data center virtualization a multi-hypervisor management problem for many companies who how have both hypervisors.
- Running applications on abstracted, shared and distributed infrastructure has driven the reinvention of the Application Performance Management business causing vendors like AppDynamics, AppEnsure, AppFirst, AppNeta, Aternity, Bluestripe, Boundary, Confio Software, Correlsense, dynaTrace, ExtraHop Networks, Gigamon, HP, INETCO, New Relic, Prelert, and Reveille Software to bring to market ground breaking new solutions that addressed new languages, were much easier to deploy, were much less expensive, and that addressed a much broader set of the application estate than the previous enterprise niche vendors were able to address.
- Agile Development created a situation where there were more new applications arriving at a faster rate than ever before, and the existing applications were changing at a rate faster than ever before. This high rate of change created a new methodology for supporting applications in production, DevOps which totally circumvented the elaborate change control procedures that were the hallmark of ITIL starting the downfall of ITIL, and all of the ITIL based management tools.
- The continuation of Moore’s Law ensured that commodity servers running cheap software overwhelmed the entire notion of scaling up into larger servers. Applications were componentized to leverage scaled out hardware creating the notion of an application that was broken in many pieces, each of which was rapidly evolving due to Agile Development, and that were then collectively deployed on hundreds and sometimes thousands of scaled out virtual or physical servers.
- Hybrid and Public Cloud offerings further distributed the execution environment so that it spanned organizational boundaries. Now if you want to manage the operations of your environment or the performance of your applications, you need tools that work across both data center boundaries and across the boundaries (and firewalls) of data centers owned by different companies. Think part of your environment running in your data center, and part running up at Amazon with pieces of applications moving back and forth an you will get the picture.
- Finally and most importantly, vendors like SolarWinds and Splunk pioneered the “easy to try and easy to buy” model of selling management software, and teams managing virtualized and cloud based systems and applications started demonstrating a strong preference for this model over the model of the enterprise sales rep wining and dining the executives who did not have to actually use the software until an ELA got signed. This new model results in two huge changes. One was that management software became much less expensive. The other and far more important change is that management software had to prove that it actually worked and provided value in the customer’s environment BEFORE the customer purchased it.
- Lastly, the most disruptive technical changes are still in front of us. Network virtualization and storage virtualization are likely to create management problems much more significant than those that were created by just virtualizing CPU and memory.
- The combination of virtualized CPU, memory, networking and storage into a Software Defined Data Center (SDDC) will make managing today’s virtualized data center look and feel like childs play.
Implications for Legacy Management Software Vendors
The combination of the new technical challenges listed above, the new environments which much be addressed and the new business models that are the basis of how customers buy management software mean that legacy management software vendors are in an extremely difficult situation:
- Their products do not work in the new world and this problem is going to get worse as network virtualization and storage virtualization gets rolled out
- They cannot acquire a new small company and glue its product to the side of their enterprise management framework to address the above issue. They have to start over with a product set that is built from the ground up to address the new requirements of the new computing environment.
- Their legacy products are being uninstalled by customers as servers get virtualized. Customers are increasingly voting with their dollars to purchase solutions designed to meet their new requirements from new vendors instead of waiting for the legacy vendors to update their products.
- The new business model for management software often cuts a zero off of the end of the price tag for the cost of a deployment. This is great for a new vendor who does not have a legacy revenue stream to protect, but it is a disaster for a legacy management vendor who faces the prospect of having to heavily invest in new technology in order to replace an existing revenue stream with a smaller one.
Implications for Enterprise Customers
One thing is certain. If and when a legacy enterprise management vendor like BMC is taken private there will not be a sudden increase in the investment in new product development. There will not be a sudden increase in the rate at which existing products are improved. And the ability for the company to make acquisitions will be severely constrained by the terms of the buyout and the amount of debt that the company will have to take on as a result of the buyout. The most likely course of action for the company is then severe cutbacks in expenses (meaning new product development), and severe cutbacks in any ability to bring any new solutions to the market. In other words, when a legacy enterprise management vendor is taken private the most likely effect is that the products will go into “maintenance mode” which is a nice way of saying that the company wants its customers to pay for maintenance every year, even though the company is investing little to nothing in keeping the product up to date. If you remember the “old CA”, where “bad software went to die”, this is the fate that awaits BMC and any other legacy management software vendor who is taken private.
This creates the following imperative for enterprise customers who are running modern data center virtualization environments (which would be just about every enterprise customer on the planet). It is time to start planning for the replacement of your legacy management software products with a new set of products built from the ground up to address the new world and the new way in which you want to try and buy software. For customers of BMC, IBM and CA it means replacing the products from those vendors with ones from newer companies whose products and business models meet modern requirements and business practices. For customers of HP it means keeping a close eye on the rapidly evolving new product lines of virtualization and cloud aware operations and applications management solutions.
Here at The Virtualization Practice we have created a Reference Architecture for your SDDC Management Stack. We refer you to the following articles:
- Building a Management Stack for Your Software Defined Data Center
- The Big Data Back End for the SDDC Management Stack
- SDDC Operations Management
- SDDC Data Protection
- SDDC Infrastructure Performance Management
- SDDC Application Performance Management
If BMC is taken private, the same fate will likely befall other legacy enterprise management software vendors like CA, and the management software divisions of IBM and HP. This will start the process of the replacement of these products by new ones built from the ground up for virtualization and cloud computing, and mark the start of the disruption of the entire management software business.