At the 2017 VMworld conference, VMware's Jose Alamo explained some of the common pitfalls associated with capacity management in SDDC and what can be done to fix it.
As companies continue to move toward the software-defined data center (SDDC), abstracting away different layers of their infrastructure, many challenges can emerge. One of the most prominent problems is capacity management.
At the 2017 VMworld conference in Las Vegas, VMware architect and consultant Jose Alamo explained how businesses can approach capacity management in the SDDC. Because of a lack of strategy, Alamo said, many companies begin their SDDC journey without a proper understanding of demand or capacity requirements, and they end up needing more capacity and potentially dealing with impacted performance.
In addition to understanding both capacity management and demand management, and how they work together, Alamo said that companies targeting SDDC should develop a core strategy for dealing with the changes in capacity that will come with SDDC implementation. Here are the six things that every company's SDDC capacity strategy must take into account in order to drive success.
1. Workload cluster distribution
How IT breaks up their workload groups can drive different needs and different policies for those workloads. Alamo said that he recommends grouping workloads by environment, but many companies still utilize mixed groups.
2. High availability
Availability is a key consideration for many IT projects, and SDDC is no exception. Alamo recommended that companies weigh the availability requirements of different environments so they'll be able to best determine what resources can be used.
3. Workload density
How dense your workloads are will also play into capacity management, Alamo said.
In order to successfully manage capacity in the SDDC, Alamo said that dashboards providing clear views into data and assets better help the business understand trends in capacity.
5. Process relationships
Admins must understand how process relationships will drive changes in capacity. For example, event management will play into incident management. This, combined with your OLAs, will have an impact on capacity. Also, seek to understand the process from end-to-end. Collect demand, compare demand and capacity, determine physical infrastructure needs, and then create and maintain a capacity plan, Alamo said.
6. Separation of roles
In order to avoid problems with crossover of responsibilities, IT leaders must seek to separate the roles and responsibilities of the team in charge of capacity management in the SDDC. Example roles could be the demand manager, demand analyst, capacity manager, capacity analyst, and capacity operations. Also, Alamo noted, RACI and roles and responsibilities documents should exist to show what each team member is responsible for.
- Software defined data centers: The smart person's guide (TechRepublic)
- How VMware paved the way for the rise of SDDC (ZDNet)
- The software-defined data center: Security is a battlefield (TechRepublic)
- Containers and their role in the Software Defined Data Center (ZDNet)
- The Ultimate DevOps Mastery Bundle (TechRepublic Academy)