The data center is home of computational power, storage, and applications that are necessary to support large and enterprise businesses. The data center infrastructure is central IT architecture, where all contents are sourced or pass through. Proper design of the data center infrastructure is precarious, and performance, scalability, and resiliency, require to be carefully considered.
Another vital aspect of the data center design is flexibility as quickly deploying and supporting new services. Designing of flexible architecture that can support new applications in very short period of time frame can result in a substantial competitive advantage. Such type of design needs compact primary planning and attentive consideration in the areas of port density, access layer uplink bandwidth, server capacity, and over-subscription, to name just a few.
The data center design is built on a supported layered approach, which has been verified and improved over the past several years in some of the major data center employments in the world. The layered methodology is the elementary foundation of the data center design that improves scalability, flexibility, performance, maintenance, and resiliency.
Data Center Types
According to architecture of data centers and the services they offer; there are mainly three types of data centers
- Traditional Data Center
- Modular Data Center
- Cloud Data Center
1. Traditional Data Center
In the traditional model of data center, there are many servers in racks, and each of them performed different tasks and applications and if redundancy required for managing critical applications, more servers are required for this task. The concept of traditional data center in the form of internet and non-stop computing is started in late 1990s where businesses and companies require a permanent presence on internet.
In traditional data center, a server is acquired, deployed and maintained individually. With the passage of time, additional servers are used with different hardware design and architecture and the data center managers are faced with learning new hardware features. In addition, the servers, operating systems such as Windows, Linux etc. applications have to be monitored and this carries different monitoring tools with additional knowledge and expertise required from the data center managers.
Updating and patching is also another concern because the updates and patches have to be verified on different hardware and software atmospheres to make sure that they will not affect the production systems. When all these costs are added, it requires high capital expenditures (CapEx) for acquiring systems and high operational expenditures (OpEx) for maintaining the structures. Even not as good as, the servers are underutilized dropping your return on investment. And lastly, the cooling and powering cost of the data center is significantly high.
2. Modular Data Center
A modular data center architecture is a portable system of deploying data center environment. A modular data center can be employed anywhere where data capacity is required.
Modular data center environments consist of purpose-engineered components and modules that offer scalable data center capability with multiple cooling and power options. Modules can be shipped to be added, integrated into an existing data center or combined into a system of modules. Modular data centers typically consist of homogeneous modules.
Modular data center structures are normally stimulated as converged infrastructure, helping in scale and efficient energy usage, including concerns regarding the external environs. A module can be treated as a single unit for U.S. Federal Communications Commission compliance certification rather than all discrete systems. Patents have been taken out on variations.
The example of modular data center systems is Sun Blackbox in 2007 – deployed with 280 servers in 20-foot container and claimed cost only 1% of cost of traditional data center.
Modular data centers normally come in two types of form factors. The more common type, stated as containerized data centers or portable modular data centers, fits data center equipment (compute/servers, storage and networking equipment) into a standard shipping container, which is then transported to a preferred location. Modular or containerized data center architecture typically built with their private cooling and chilling systems. Cisco makes an example of these type of data centers, called the Cisco Containerized Data Center.
Another practice of modular data center fits data center equipment into a capacity consists of prefabricated components that can be quickly constructed on a site and added to as capacity or additional capacity is required. For example, HP’s modular data center architecture is formed with sheet metal components that are shaped into four data center halls linked by a central operating building.
3. Cloud Data Center
Cloud computing or cloud data center ensures more power, safer data, and easier access to the information and data and tools required for success in any business or organization. Cloud provides a comprehensive array of IT infrastructure and services for migrating smooth and cost-effective into the era of cloud computing — end-to-end solutions that ensure efficiency, agility and honesty for your users, data, and applications.
Cloud Computing offers a range of cost-effective, well-organized, and flexible cloud data center solutions. These powerful cloud solutions are constructed upon demonstrated core technologies with an open design that integrates software products and high-performance hardware:
- Virtualization of compute, storage, and networking resources
- Software-Defined Networking (SDN) competencies
- Cloud management tools for unified resource operation, automation, management, and maintenance
- Service and business continuity with disaster recovery and prevention
The Future of Cloud Computing
There are many potential opportunities and capabilities with cloud computing and the technology behind it. Cloud computing can open a complete new realm of opportunities, services, platforms, applications, and even more. There are hundreds and thousands of opportunities beginning to form as the future of cloud computing starts to really take off.
Michael Corrado, World Wide Marketing Manager with Hewlett Packard Enterprise says, “The future of cloud computing will most likely represent a combination of cloud based software products and on premises compute to create a hybrid IT solution that balances the scalability and flexibility associated with cloud and the security and control of a private data center.
In the current cloud market, the benefits of leveraging the infrastructure of a large cloud provider can be beneficial in many ways. The cost structure works like a utility which provides for an operating expense model with no upfront infrastructure costs.”
According to Forbes.com report released in Apr, 2017, Cloud computing is projected to increase from $67B in 2015 to $162B in 2020 achieving an annual growth rate (CAGR) of 19%. Cloud platforms are empowering new, complex business models and orchestrating more globally-based integration networks in 2017 than many technical analysts and advisory firms predicted. Increased Cloud Services adoption in mid-tier and SMBs, leading researchers such as Forrester are adjusting their forecasts upward. The best form of check of any forecast is revenue. According to Amazon’s latest quarterly results show that AWS attained 43% year-over-year growth, contributing 10% of consolidated revenue and 89% of consolidated operating revenue (Columbus, 2017).
According to Forbes.com in Feb, 2017, Global cloud IT market revenue is predicted to increase from $180B in 2015 to $390B in 2020, attaining a Compound Annual Growth Rate (CAGR) of 17%. Global cloud market revenue is predicted to increase from $180B in 2015 to $390B in 2020, attaining a Compound Annual Growth Rate (CAGR) of 17%. In the same period, SaaS-based apps are predicted to grow at an 18% CAGR, and IaaS/PaaS is predicted to grow at a 27% CAGR (Columbus, Global Cloud Predicted To Reach $390B By 2020, 2017).
As cloud requirements and adoption are increasing day by day, trained manpower is also required to manage cloud platforms either it public, private, hybrid, or a multi-cloud environment. To build a cloud architecture, it should have:
- Heterogeneous support system that can leverage the latest hardware, virtualization, and software solution, but they should also support existing infrastructure of a data center.
- Service management system where an administrator should have a simple tool for measuring service offerings to productize the cloud computing environment.
- To be dynamic in Resource management and workload where a cloud can be truly on-demand and elastic.
- Reliable, available, and secure
- Integrate with data center management tools by using Application Programming Interfaces (APIs).
- Complete visibility and reporting in performance and service level perspective
- Provide interfaces for administrator, developer, and end-user
According to Forbes report, revenue is predicted to increase from $180B in 2015 to $390B in 2020, and as per predictions, 60% of all IT jobs will be cloud based by 2020. Due to heavy spending and adoption of cloud computing, demand of developers in Cloud, SDN, and NFV is increasing. If you are interested in building cloud applications and their automation, you will have to learn and excel in gaining experience in one or more tools mentioned above. In this article, We’ve discussed future of cloud computing, its architecture, role of developers in cloud computing, and some important tools and scripting languages which are helpful for building cloud applications and automating the environment.
Columbus, L. (2017, February 11). Global Cloud Predicted To Reach $390B By 2020. Retrieved from Forbes: https://www.forbes.com/sites/louiscolumbus/2017/02/11/global-cloud-spending-predicted-to-reach-390b-by-2020/#60e83da21085
Columbus, L. (2017, April 29). Roundup Of Cloud Computing Forecasts, 2017. Retrieved from Forbes: https://www.forbes.com/sites/louiscolumbus/2017/04/29/roundup-of-cloud-computing-forecasts-2017/#41481b5331e8