Benefits and Limitations of Data Center, Cloud, and Colocation Solutions
NTT Global Data Centers Americas’ CHI data center in Itasca, Illinois. (Photo: NTT Global Data Centers Americas)
Last week, we continued our series of special hybrid cloud reports. This week, we’ll take a look at the benefits and limitations of data center, cloud, and colocation solutions.
Businesses have a range of options for their IT needs. Data center, cloud-only, and colocation solutions all have significant advantages and limitations. One âsizeâ does not fit all the needs of today and the rapidly changing demands of tomorrow.
Data center: the anchor
Traditional data centers offer many advantages. They hold and store institutional knowledge of the company’s IT policies and philosophies within its staff, documentation, and equipment and typically serve as the coordinating organization for all IT resources including cloud and colocation. .
- Company-owned space and equipment provide high levels of data control
- Servers operated and maintained by employee corporate badges
- Physical access control for servers and other computing resources may not be as strict or intrusive as the multiple layers of physical security found in colocation facilities.
- Data centers can provide IT support to the business
- User-oriented frontline help desk organizations
- Second level support for more complex issues and issues
Expansion is very costly in capital. Existing installations cannot be shut down mid-operation to be retrofitted for improved power and cooling to meet compute-intensive workload demands.
- The data center environment is designed for operations and stability
- Not designed for wholesale experimentation, development, scaling / hyperscaling
- Difficult to sandbox and deploy publicly accessible applications within the security perimeter of corporate IT operations.
- Expansion is very expensive in capital
- Difficult to be agile with new projects and installations
- Existing installations cannot be shut down mid-operation to be retrofitted for improved power and cooling to meet compute-intensive workload demands
- Network connectivity is only as good as provisioned
- Upstream broadband outages affect the data center and the entire enterprise, as branch office data traffic is routed through a central location for management and security.
Cloud: easy, scalability and customization limits
The idea of ââbuying IT as a service is not new, with today’s cloud offerings simplifying the process into a simple e-commerce transaction.
- Ease of purchasing via web portal and credit card
- Many services available through the cloud
- Software, storage, backup, APIs integrated with larger business applications, bare metal servers, virtual machines, container environments and GPU services for machine learning and other specialized tasks
- Enterprise developers can easily “sandbox” applications outside the corporate security perimeter
- Little / no risk for critical data
- Applications and resources scalable to some extent
- Scalability available without overloading existing data center infrastructure
- Cloud services have single points of failure
- At the point of application service provider
- Upstream at the cloud provider
- Third-party control and operation of the cloud can lead to higher security risks
- Both application service provider and cloud provider are attack surfaces
- Pure size makes clouds lucrative targets for bad actors
- Cloud services do not offer the tailor-made customization needed for specific issues
- Cloud optimized for volume delivery based on least cost rather than best service delivery
- Clouds do not economically adapt to big problems
- More expensive than a data center or colocation using an optimized solution with dedicated hardware.
Colocation: where retail and hyperscale converge
With growing needs for high performance computing in all applications, the worlds of large-scale computing and retail colocation are starting to merge. High-end colocation facilities allow businesses to build and control bespoke facilities designed for their large-scale needs, with the capacity to support 5,000 or more servers under one roof, from network connections to high speeds. of 40 Gbps and above, and the ability to deliver megawatt-class power to support GPUs and other dense, power-hungry compute configurations not designed for the typical retail data center scenario.
Using highly tuned systems with servers optimized for e-commerce and other low latency, high demand workloads, GPUs for AI / ML, and data analytics applications running with large Amounts of storage are among the factors pushing retail colocation into a large-scale territory as businesses realize they need dedicated computing resources for difficult issues beyond the cloud economy.
The benefits that high-end colocation facilities bring to hyperscaling include
- The ability to create a customizable hardware and network architecture to scale, leveraging in-place physical security, HVAC power and cooling
- Availability of megawatt class power for dense computing needs such as GPUs
- Superior network connectivity, including
- High-speed, low-latency connections to carriers and major exchange points with multiple physical paths for redundancy
- Access to dark fiber for direct connections between the company and colocation facilities
- Meet-me-style neutral network exchanges allowing Gigabit Ethernet connectivity to multiple carrier networks and hyperscale clouds as needed
- Facilities can be physically located in specific regions or countries, ensuring that data is not removed from the originating country / region due to regulatory requirements
Connectivity comes into play at two different levels to build these solutions. The availability of dark fiber to directly connect the business to the colocation facility is likely to be a must in many cases for both security and low latency. If workloads need to be more widely accessible outside the enterprise by customers and partners, network connectivity through Meet-me network exchanges ensures low latency access for users.
Download the full Hybrid Cloud report, courtesy of NTT to learn more about how workloads continue to switch between datacenter, cloud, and colocation. In our next article, we’ll see how workloads determine current and future compute needs. Find previous articles here and here.