|By Stefan Bernbo||
|August 7, 2013 10:30 AM EDT||
Cloud computing has opened the doors to a vast array of online services. With the emergence of new cloud technologies, both public and private companies are seeing increases in performance gains, elasticity and convenience. However, maintaining a competitive advantage has become increasingly difficult. Service providers are taking a closer look at their data storage infrastructure for ways to improve performance and cut costs.
If the status quo remains, maintaining low-cost cloud services will become increasingly difficult. Service providers will incur higher costs, while consumers become burdened with storage capacity restrictions. Such obstacles are influencing service providers to find new ways to scale cost-effectively and increase performance in the data center.
In response to the increase of online account activity, service providers are consolidating their data centers to a centralized environment. By doing so, they are able to cut costs while increasing efficiency, allowing data to be accessible from any location. Centralizing equipment enables providers the ability to deliver enhanced Internet connections, performance and reliability.
However, with these added benefits also come disadvantages. For instance, scalability becomes more expensive and difficult to achieve. Improving efficiency within a centralized data center requires the purchase of additional high-performance, specialized equipment, which increases costs and energy consumption, challenging endeavors to control at scale. In an economy where cost-cutting is becoming a necessity for large and small enterprises alike, these added expenses are unacceptable.
Characteristics of the Cloud
Solving performance problems, like data bottlenecks, is a growing concern for cloud providers who must oversee significantly more users and accompanying performance demands, than do enterprises. Although the average user of an enterprise system requires elevated performance, these systems generally manage fewer users who are able to access their data directly through the network. Moreover, enterprise system users are accessing, saving and sending comparatively relatively small files that require less storage capacity and performance.
Outside the internal enterprise network, however, it's a different story. Cloud systems are simultaneously being accessed by a multitude of users across the Internet, which itself becomes a performance bottleneck. The average cloud user stores relatively larger files than the average enterprise user placing greater strains on data center resources. The cloud provider's storage system not only has to scale to each user, but must also sustain performance across all users as well.
In response to growing storage demands, cloud providers are faced with profound business implications. Service providers need to scale quickly in order to meet the booming demand for more data storage. The following best practices can help optimize data center ROI in a period of significant IT cutbacks:
- Opt for commodity components when possible: Low-energy hardware makes good business sense. Commodity hardware is not only cost-effective, but also energy-efficient, which significantly reduces both setup and operating costs in one move.
- Seek out a distributed storage system: Distributed storage presents the best way to build at scale even though the data center trend has been moving toward centralization. Increased performance at the software level counterbalances the performance advantage of a centralized data storage approach.
- Avoid bottlenecks: A single point of entry can easily lead to a performance bottleneck. Adding caches to relieve the bottleneck, as most data center infrastructures currently do, quickly adds cost and complexity to a system. On the other hand, a horizontally scalable system that distributes data among all nodes delivers a high level of redundancy.
Currently, Big Data storage consists mainly of high performance, vertically scaled storage systems. Since these infrastructures can only scale to a single petabyte and are costly, they are not a sustainable solution. Moving to a horizontally scaled data storage model that distributes data evenly onto energy-efficient hardware can reduce costs and increase performance in the cloud. With these insights, cloud service providers can take the necessary steps to improve the efficiency, scalability and performance of their data storage centers.