Scale-out architecture: hyper scalability begins at home
Bringing the concepts of scale-out architecture to data management and secondary storage
“All in.” It’s a great phrase when looking at enterprises’ use of the cloud versus on-premises for data management and secondary data storage. Are enterprises going all in on the cloud for all their data protection and workload needs? Or, are they seeing that while the public cloud, for example, might be optimum for Tier 2 applications, they prefer to retain certain critical data processes on-premises? As Gartner noted, many companies are not “all in,” but their journey to the cloud is a slow, controlled process.1
Scale-out architecture plays in this scenario, as IT executives are starting to see it can offer the virtually unlimited scalability enterprises associate with the public cloud, however, with tighter budget control. This brief will help you examine where scale-out architecture can be effective in supporting increased data workloads, and providing agility, while optimizing your hardware expenses.
The public cloud is offering enterprises the convenience of scalability and easy availability but IT executives are learning painfully well the public cloud can also come with a high price tag. Cloud data migration is one of the drivers to consider as IT executives look at on-premises operations to see where they can achieve scalability and flexibility, but with a controlled budget to maximize business value.
In this brief, we will suggest questions you will want to ask yourself and your teams in considering deployment of scale-out, data protection and secondary storage architecture. This includes transitioning from traditional, single-purpose scale-up architecture to economical, general purpose, hardware that can support applications and workloads with cloud-level flexibility.
The big data picture
Change is good. But enterprise change is better when it syncs with the overall rationale of what you want to achieve in modernization of your IT infrastructure.
Before making a determination on deploying scale-out architecture for functions such as secondary storage, data protection and management, collaborate with your team on these strategic questions:
- Are you all in on the cloud? If your longer-term strategy for IT architecture includes retaining designated processes on-premises, or specific applications cannot, or should not, be moved to the cloud due to prohibitive costs, scale-out architecture becomes a viable option for secondary storage and data management. It can serve as a means of achieving cloud-like agility, yet enabling your IT staff to more effectively utilize existing resources and to more closely follow budget parameters.
- Where are you on the virtualization path? Is your organization still using traditional build-your-own backup servers or are you more fully deployed in virtual servers? This is a good time to consider leveraging the concepts of hyper-converged infrastructure and applying them to data management and secondary storage. Using scale-out architecture, you can provide an integrated solution that supports hardware, software and virtualization. IT can benefit from time savings and reduced operational complexity.
- Can you call yourself a digital business yet? Around the corner from virtualization is the agility of the digital business. Where are you in terms of embracing digital technologies to become more competitive? Have you made the transition to a data driven enterprise as suggested by Forbes? “As data emerge and the ability to associate that data with indicators or issues, it enables an organization to not only become more efficient but also change what it does. The business impact resulting from viewing the world through a data lens instead of a process lens is super-powerful.”
Data-driven organizations need the scalability and agility that scale-out architecture can provide. Secondary storage and data protection must be managed at the competitive speed enterprises are demanding to ensure the resiliency and availability of a true digital business.
Getting an ‘A’ in operations
IT executives know painfully well all new ideas, and technological implementation, must come with a business value justification. That goes as well for moving to scale-out architecture in order to deliver scalable and flexible services on-premises. Since operational excellence is a core tenet of IT business value, you need to determine where scale-out secondary storage and data protection can add the most value.
Here are some questions to consider:
- Are you seeing the need to reduce operational complexity due to growing storage capacity along with data growth?
- Has the total cost of ownership (TCO) of traditional data protection become excessive? Are you therefore looking to reduce the TCO of secondary storage and backup environments?
- Are you finding that specific applications are not supported in the cloud, are too costly to migrate to the cloud, or must reside on-premises due to regulatory or compliance requirements?
- Is moving data between storage environments, and between clouds, becoming increasingly inefficient and difficult?
- Are internal users complaining about recovery point objectives (RPO), and are recovery time objectives (RTO) no longer meeting their needs? Are service level agreement (SLA) performance metrics slipping?
Become a singular sensation
You have evaluated scale-out architecture against strategic initiatives, business value and performance metrics. Your team agrees scale-out is a good fit to deliver greater resiliency, availability, and scalable services on-premises. Now it’s time to consider how you can show value by being a singular sensation – leveraging the benefits of a single software-defined stack to deploy a scale-out infrastructure.
You will want to evaluate an infrastructure solution that provides complete hardware (compute, storage, and network) and software (single application, OS, and storage abstraction) in a single solution. This single platform should be supported by a unified management interface that reduces TCO, and provides a single, complete view of all enterprise data wherever it resides, eliminating data silos and point solutions.
Another aspect to evaluate is using scale-out hardware to support applications, databases, VMs, and cloud needed by enterprises to successfully deliver data management services to their end-users. As RedHat defines software-defined storage (SDS) as a storage architecture that separates storage software from its hardware. SDS is designed to perform on any industry-standard or x86 system, removing the software’s dependence on proprietary hardware.” 3 By being able to use commodity x86 hardware, your IT teams can realize dramatic cost savings, and by scaling on demand, you also avoid vendor lock-in and unnecessary costs.
Scale-out begins at home
While enterprises continue to migrate data to the public cloud, most will continue to execute some secondary storage and data protection and management functions on-premises, or in their data home. Scale-out architecture is the perfect bridge to this future, providing cloud-like data protection on-premises, yet with the scalability and speed digital businesses demand.
Take the time to work with your teams to see whether using scale-out architecture for secondary workloads, including data management and protection, is a good first step in upgrading your overall infrastructure to one that is cloud-agile, yet budget controlled.