There are tons of good reasons to venture to the cloud, specifically like the demands on businesses and their data centers to be flexible. As many of you know by now, cloud offers reduced costs, simpler implementation for future products, and an overall easy to maintain and agile environment.
In a recent study by IDG Research Services, more than 100 IT executives were surveyed and 48% of these respondents shared that they were in the evaluation, consideration, or planning stage. Many of these IT execs are desperately trying to do more with less, and cloud is the right remedy – allowing users to provision storage and network resources in a fraction of the time compared to traditional and virtual servers.
However, many of these in the planning stage get in too deep before they realize migrating their data centers is not as simple as predicted. In fact, only 15% of the respondents said they were “very successful” in designing their infrastructure.
So Why Does Infrastructure Matter So Much?
Take the time during the design phase to create a logically integrated, centrally managed infrastructure that will support your migration. Silo-based design and deployment eats up your budget and timeline, as well as reduces the expected cloud benefits.
Companies that ignore the need for a thorough design phase experience negative outcomes. For example, in the same IDG Survey, 55% reported increased costs, 41% reported slower implementation times, and 38% increased their power and cooling needs.
Agility, one of cloud’s biggest selling points, is obsolete without a unified physical infrastructure. Why? Because handling physical moves, adds, and changes is painful and suffocates the demanded flexibility.
Benefits of a Unified Physical Infrastructure
The top benefits of deploying a cloud solution with a well thought out infrastructure include:
- Minimizing unplanned outages
- Reducing infrastructure management costs
- Increasing time available for strategic activities and faster provisioning
- Reduced infrastructure complexity
- Better operational control
- Improved confidence in layer 1 stability
- Documented and approved provisioning process
Infrastructure Questions and Considerations for Private Cloud
According to a white paper released by Panduit entitled “Physical Infrastructure: a Critical Factor in Cloud Deployment Success”, there are five categories, each with critical questions you should consider before deploying a private cloud.
1. Increased Server Bandwidth Requirements
- What demands will virtualization place on the speed and bandwidth of your network?
- By reducing the number of physical servers, the number of virtual machines on each physical server increases (as does network bandwidth requirements on the server). What would increased bandwidth on your existing hardware and network infrastructure mean to you?
2. Managing Physical Server I/O Proliferation
- Many companies experience a large increase in the I/O per server as they adopt virtualization. How will you manage high cable densities?
- How has virtualization changed the way you architect your network and IT infrastructure?
3. Need to Efficiently Plan Cabling Infrastructure
- Virtualization leads to an increase in bandwidth and number of cables connected to each server. How will you manage up to three times more cables in each cabinet and in your pathways?
4. Need to Plan for Greater Power and Cooling Needs
- What steps are you taking to increase the cooling efficiency of your data center?
- How does your team monitor and manage the available power and cooling for a given server or cabinet? Would it be valuable to have the ability to remotely manage connectivity, power and environmental sensing of network equipment? Do you have initiatives to improve mean time between failures (MTBF)?
5. Cloud-in-a-Box Impact on the Data Center
- Integrating cloud-in-a-box into the data center infrastructure can be a challenge. Traditionally, enterprise data centers do not mix server, storage, and network devices in a single cabinet. How will your team account for the thermal management of active equipment with different cooling requirements in a single cabinet?
- How will your team ensure your data center is ready for a cloud-in-a-box deployment?
- What is your plan for weight distribution and power balancing per power outlet and unit of equipment when creating or implementing cloud-in-a-box?
Infrastructure Questions and Considerations for Public Cloud
For the public cloud, there are four categories which you should consider before deploying a public cloud.
1. Fast Implementation to Accommodate Rapid Growth:
- How do you streamline the process of designing, specifying, installing and managing the increasingly complex physical infrastructure necessary to optimize your data center?
- How do you facilitate faster implementation of the physical infrastructure in your data center?
2. High Density Server Deployments Exceed Available Power and Cooling Supply:
- How does your team monitor and manage the available power and cooling for a given server or cabinet?
- What value would remote management of connectivity, power and environmental sensing of network equipment bring?
3. Maximize Revenue Generation per Square Foot or Kilowatt:
- How do you ensure that you maximize revenue in your facility? This includes maximizing floor space and vertical space in the cabinet.
- Cable congestion in cabinets can cause cooling issues due to blocked airflow, and can complicate moves, adds and changes, increasing costs and risk. How will you manage high densities of cables?
4. Ability to Meet or Exceed SLAs:
- How do physical infrastructure moves, adds and changes impact your ability to meet your SLAs?
In Conclusion…
If you haven’t done so already, don’t be afraid to ask for help. 59% of those very IT execs surveyed admitted they rely on or would rely on outside help for design and implementation. Many things are too good to be true, especially when improperly managed. Whether you’re migrating to a public or private cloud, the earlier you plan for infrastructure, the quicker you’ll see a return on investment.
