The converged architecture model has a number of benefits. You can reduce your overall data center footprint, support more business use cases, create greater levels of agility, and generally support the business in dealing with today’s dynamic market changes.
With all of that said, when deploying converged architecture, you have to take design and resource utilization into consideration. Are you deploying one pod for a specific use case? Or maybe you need HCI and require multiple clusters to support your business. Whatever the case, sizing and design are important considerations when deploying converged systems.
With that in mind, here are five points to consider when designing a converged architecture solution:
1. User count
Administrators must know how many users will be accessing the environment at a given time. Even more important will be their ability to forecast the user count a year or two down the road. This type of planning will help dictate the size of the converged architecture platform and how many converged blocks will be necessary. By establishing user count and future growth metrics, you’ll also be able to tell how much storage (IOPS), RAM and CPU are necessary. From there, it’s critical to understand how you’ll be segmenting these users. Remember, converged architecture supports powerful multitenancy options. User count and user definition will allow you to properly segment user workloads, business units, and even specific virtual desktops and apps.
2. Workload type
What are you going to be delivering? Are these virtual desktops or just published applications? Are you supporting a mobile user, or are many users locally based? Will users be working with resource-intensive workloads? All of these questions will help determine the type of converged architecture necessary for your use cases. By knowing the workload type, you can size and plan the environment much more effectively. Furthermore, you can wrap optimization policies around a specific workload. And, from a control perspective, you can segment specific workloads for compliance, security and/or optimization purposes.
3. WAN link
In a distributed converged architecture environment, connectivity is critical. It’s important to understand what we are pushing down the link to ensure optimal performance for the end user. Are you delivering rich media content or just small files? In addition, bandwidth must be considered when the converged architecture platform is being designed—not after. Remember: With CA you’re integrating storage, compute and network resources. Many organizations are supporting distributed environments in which network integration plays a critical role in workload delivery. Know your WAN and local network requirements when creating converged architecture solutions. This will help you remove network bottlenecks and allow for the proper delivery of your workloads.
4. Data center considerations
Before you deploy a converged architecture solution, it’s critical to examine your existing data center environment. Do you have enough space? What are your power and cooling requirements? Are you in need of a redesign? CA platforms are deployed as validated architectures designed to run efficiently and reduce data center space. When you examine your existing data center, determine where CI will fit in and where it will create efficiencies for your overall business. Are you removing older gear? Can you optimize HVAC requirements? Converged architecture helps deliver powerful workloads, but it can also help you run a more efficient data center.
5. Delivery methodologies
The success of a virtualization-ready, distributed CA environment will largely depend on the acceptance level of the user base. With poor performance and a fluctuating experience, satisfaction and adoption rates will be low--leading to the possible failure of the entire project. When it comes to converged architecture, part of capacity planning must be how the workload is delivered down to the end user. What speeds are optimal? Where will certain types of content be rendered? Do adjustments need to be made to compensate for latency? All of these questions must be answered prior to deploying a production converged infrastructure system. Remember, virtualization plays a big role within a CA implementation. You’re enabling multitenancy on physical and virtual layers. When the delivery methodology is properly identified, you can wrap appropriate policies around the physical CA environment.
When working with converged architecture, sizing and design are critical concerns. One of the biggest benefits of CA is the ability to design and deploy preconfigured and validated architectures, which speeds the deployment of virtual machines by creating an architecture that simply “snaps” into place. With thoughtful upfront design, you can develop true data center optimization that will not only support but also drive your business.
Underwritten by HPE
Part of HPE’s Power of One strategy, HPE Converged Architecture 700 delivers infrastructure as one integrated stack. HPE Converged Architecture 700 delivers proven, repeatable building blocks of infrastructure maintained by one management platform (HPE OneView), built and delivered exclusively by qualified HPE Channel Partners. This methodology saves considerable time and resources, compared to the do-it-yourself (DIY) approach.
Based on a complete HPE stack consisting of HPE BladeSystem with Intel® Xeon® E5 v3-based HPE ProLiant BL460c Gen9 blades, HPE 3PAR StoreServ all-flash storage, HPE Networking, and HPE OneView infrastructure management software, the HPE Converged Architecture 700 can be easily modified to fit within your existing IT environment.