Tips for choosing the best offering to bring efficiency and agility into your business
This is the third blog in a series on Converged Infrastructure. Read the previous blog post here.
I'm closing in on my two-month mark here at Dell. I must admit, it's been one heck of a ride so far. There are several critical observations I've found while talking with customers and other members of my team regarding infrastructure convergence — how it is currently being implemented and used, and, ultimately, where it needs to head moving forward.
The greatest realization as I've settled into my role here is how intertwined the hardware and software management is to the entire concept of Converged Infrastructure. Convergence helps organizations achieve levels of agility and efficiency that simply haven't been seen in a data center since the mainframe days. But, I'm quickly finding that the hardware can only enable a customer to gain these agility and efficiency benefits of convergence if the software layer managing the hardware stack enables a certain level of Cross-IT simplification. We are currently seeing gaps in the market around a complete solution from any one vendor.
The first reason for this is the proprietary nature of existing converged infrastructures. Vendors are trying to lock customers in with proprietary hardware and software elements, which, ultimately, may not even be complete solutions or are only designed to work for specific use cases. Deviation from these intended scenarios can be costly and require additional investment from adopting customers. The reason for this lock-in is simple: these vendors want to protect and expand their existing footprint in customer environments. It's simply smart business for a company to focus on their strength.
Customers do need to be aware of when vendors require proprietary hardware and push customers towards costly CapEx situations when entire infrastructure elements need to be replaced or deviated from a corporate standard. This also has an OpEx impact as admins learn new platforms and how to manage new, and often complex, components.
The second reason I have found is that existing solutions are incomplete, and are simply lacking a hardware infrastructure element (such as storage or support for physical infrastructure), and force customers to leverage multiple tools for management of various hardware stacks, which fundamentally goes against the core benefits of convergence. Other solutions in the market have only basic capabilities around hardware configuration and management, and force customers to look towards third-party partners to provide stronger operational management capabilities, again with significant CapEx and OpEx costs tacked on.
As soon as an organization starts to piece together converged elements across different vendors or have to use multiple management tools, things get complicated quickly. Customers want to avoid the "Vendor Lock-In" problem, yet at the same time, want the agility and IT efficiency that can only be provided through close collaboration between the server, network, storage, virtualization, and management stacks. This is extremely difficult to achieve when piecing together elements from 3+ different vendors. The shift away from IT silos is also vital, as many organizations have internal battles around which core infrastructure element is most important (server platform vs.
storage vs. network vs. virtualization), which could lead to a converged infrastructure vendor that the collective organization may regret later due to cost, complexity or an inability to centrally manage the data center of the future.
Key characteristics to focus on when choosing a solution
When beginning the enlightening path towards data center convergence, organizations need to focus on a few key characteristics of the solutions they are looking at to enhance their agility and efficiency.
First is to find a solution that is flexible. It simply means the ability to implement a solution that is interoperable with existing infrastructure investment and doesn't force customers into proprietary hardware decisions or to completely replace core infrastructure elements. It should not be necessary for an organization to have to change their storage, network, server or virtualization platform to adopt a converged infrastructure. While there may be significant benefits to aligning with a single partner, these benefits should absolutely be weighed against the cost and complexity of getting to the final destination.
The second characteristic is to look for an intuitive solution. This is primarily focused on the management of the collective stack. Look for a solution that is simple and allows admins to perform their daily tasks quickly and easily. It may not matter if a particular management tool has 98% coverage through the UI vs.Command Line if admins only required 10% of that functionality regularly, and that 10% ends up getting buried in the complexity of a bloated UI. It's no secret that I'm a HUGE fan of functional design. I've done several personal blog posts on the topic, and feel that if a vendor doesn't put emphasis on simplicity and user design, they struggle to connect with users and fundamentally understanding the pressing issues that their customers are facing daily,
Overall, the path to convergence, what to look for, what to avoid, who to buy from, how it will impact business, and a million other questions is not simple. Customers need to do their research, understand the good, the bad, and the ugly in the market, and forge a solid path with their business goals in mind.