Users can use object storage and business hardware to enter cloud computing services, but this is not the best solution. Using object storage would make sense in the context of storage requirements that support applications under the new cloud architecture that are written based on the RESTful application programming interface. But for today's applications, customers prefer other solutions, leveraging publicly available cloud storage services, cloud storage gateways and third-party products that tightly federate one or more public clouds.
How to determine what kind of workload and use cases are suitable for public cloud storage? While cloud computing development may seem attractive from an expense and scalability perspective, cloud computing is more efficient and easier than any other solution for specific types of workloads and storage use cases.
Let's first understand a bit of background. This solution has seen tremendous growth since Taneja Group first proposed a public cloud storage solution for enterprise customers in 2008. Although Amazon remains the leader in this area, today's customers have the option of more vendors and storage options to pursue the best value for money. Constant innovation and escalating competition make cloud storage more attractive than ever.
Today's mainstream public cloud vendors serving the enterprise market tend to offer three basic storage services: block storage services with tightly coupled cloud platforms (such as Amazon Elastic Block Storage [EBS], HP Cloud Storage, Windows Azure Binary (For example, Amazon Simple Storage Services, HP Cloud Object Storage Services, Windows Azure Binary Storage Services); content delivery network services such as Amazon CloudFront, HP Cloud CDN, and Windows Azure CDN. Object storage solutions are the main points of interest when developers write cloud applications based on next-generation frameworks, while block storage solutions are mainly designed to meet the aging storage requirements of today's applications.
At the same time, public cloud storage vendors often provide additional storage-related services to meet the needs of specific use cases such as relational and non-relational database applications, database synchronization, long-term archive storage such as Amazon Glacier, and cloud storage gateways such as AWS Storage or Microsoft StorSimple Cloud Integrated Enterprise Storage). There are also a range of third-party cloud backup products to choose from.
In this context, workload and storage use cases that best suit the cloud environment will be largely governed by the application and service level requirements. This is useful from a planning and evaluation point of view, which can limit your need for storage based on two key dimensions: the value / sensitivity of your data, and the nature of your storage use cases and workloads.
In short, the greater the data value / sensitivity, the higher the need for security and compliance. While public cloud security has steadily increased, this does not address the storage needs for highly regulated data that must meet tight security, privacy, and / or other compliance standards. Information such as covering sensitive medical, financial and other information are subject to industry regulations such as the US Health Insurance Portability and Accountability Act and the Payment Card Industry Data Security Standard. Many types of federal and state government data are also in this controlled directory. If your data can not afford even the slightest security threats, unauthorized access or data loss, then it can not exist in the public cloud.
Public cloud storage applications often fail to meet the performance levels and availability required by manufacturing applications, and as a result, vendors are reluctant to include in their service level terms any obligation to exceed minimum availability (and without any performance guarantees). Such as AWS guaranteeing that its Elastic Computing Cloud, Elastic Block Storage (EBS) infrastructure has 99.95% availability (that is, 21 minutes of downtime per month), but such guarantees are still available even if the client's workload is down Satisfied, because the definition of availability is that at least one instance can respond to external connections while at least one mounted disk volume can be read and write and input and output.
The best domain for public cloud storage is data value / sensitivity is very low, and focus on the field of two or three use cases, such as development / testing, backup, disaster recovery, archiving and so on. Public cloud storage provides the ideal solution in these scenarios, as storage performance, availability, and security requirements are relatively relaxed. In spite of this, customers can still take advantage of the great scalability and compelling costs of cloud computing.