Wired published “8 Steps to Building a Modern Datacenter” highlighting features of modern data centers. Companies are trying to work hard with a consistent need for storage and computer resources. The Wired article focuses on the "what" and not the "how".
- There is a greater need for compatible frameworks and consoles across network, server and storage silos.
- Modular design is simple and flexible.
- Much smaller and compact single rack solutions.
- Incorporate fixed, vendor-defined ratios for computing resources and storage capacity.
- The IT hardware has also gotten much smaller while delivering the required computing power, over the last 2-3 years.
Converge When Possible
- Fewer dedicated resources.
- Economical and efficient.
- Functions can be combined into one appliance.
- Improves scalability.
- No need to spend on additional hardware or high-speed, dedicated networking equipment.
- The days of expensive, specialized hardware in datacenters are ending.
- New services without adding hardware.
- Save on cost.
- Offer agility.
- Distributed applications.
- Improve uptime, global scalability and service continuity during site failures.
Embrace Commodity Hardware
- Scale fast with minimum investment.
- Distributed software layer abstracts all resources across clusters of commodity nodes.
- Deliver aggregate capacity.
- Surpass the most powerful monolithic approaches.
Empower End Users
- Datacenters need to be resilient and reliable.
- Meet the growing demands.
- Applications ranging from virtual desktop infrastructures (VDI) to employees handheld devices.
- Admins are moving to end-user computing models.
- Desktops, applications and data are centralized within the datacenter and accessed by employees via any device from anywhere.
Break Down Silos
- Don’t need separate teams of specialists for each technology.
- Integrating the technologies into a single scalable unit, or datacenter building block, reduces the need for specialized staff.
- Many enterprises want to use the public cloud.
- Keep business-critical applications involving confidential data safe within the confines of the private datacenter.
- Control over of security, performance and service level agreements.
Focus on Service Continuity
- Enterprise disaster recovery strategies tend to be reactive.
- Consumerization has altered user expectations.
- Admins have to be proactive and focus on service continuity.
- Enterprises should re-architect their applications to be distributed.
- Better scale globally and perform well and increase uptime.
Many start the design process with a specific solution in mind and back-chain to see if it fits the requirements. A better approach is to define the requirements and look at possible solutions. Requirements should be established. The design parameters can then be set (i.e. location, size, power density, efficiency etc.) Engineering specialists would thus be appointed.