What is Cloud Data Management?
In This Article
Data is the prized possession of every organization in the market. Whether your business was born in the cloud or you are a traditional on-premises business developing a cloud strategy, how you manage your data should be one of your primary focuses. From an application standpoint, your application cannot operate without access to its data, and every single company has mission-critical applications deployed or developed in order to operate. Coupled with performance requirements, management, security, governance, compliance requirements, and recoverability, it is easy to see that data management can be difficult without defining the proper cloud data management platform and strategy. This is where WWT can help discover, design, advise, and deploy the proper data management solution to meet your application's needs.
Many in the technology industry would define cloud data management as a platform, although there is much more to it than simply launching a platform. The platform can help with the management aspects, but you still need to define a strategy around where your data should live. This could be on-premises in a high-performance storage array, in any of the public cloud providers' myriad of storage offerings, or a mixture of on-premises and public or private clouds. We define cloud data management as "the ability to manage data across public or private cloud platforms while maintaining the ability to leverage on-premises data where necessary, in a true heterogeneous approach, while maintaining the proper level of performance, security, governance, and accessibility for your business". Cloud data management should be a strategy and best practice for all organizations to customize and adhere to as they would any other critical business operation.
With most things in life, having a plan in place before starting down a path is preferred. Managing the location of your enterprise's data is no different. Gathering key stakeholders from security, storage, and data governance departments, to name a few, can help create a holistic strategy.
As you venture out with a cloud strategy, whether this is a data migration or cloud-native approach, you must ensure that as data is moved into the cloud you can have the same availability as you would on-premises. This includes high availability SLAs, redundant architectures, and seamless failover and giveback processes. Additionally, these need to be perpetuated to the applications or hosts, as the data storage is just one piece of the puzzle.
The move to the cloud requires a shared responsibility model, meaning that you no longer have direct control over the physical hardware that is storing your mission-critical data. However, you are responsible for protecting your data and cloud resources from being deleted, becoming corrupted, or withstanding a natural disaster. Utilizing the software and cloud data management platform is key to ensure you can access your data should a failure occur, while also meeting internal recovery point objectives (RPO) and recovery time objectives (RTO) for your applications and their data.
The terms backup and archiving are critical to any enterprise. As we expand the concepts of backups or archiving to the cloud, away from traditional tape, a single framework for performing these backups and archives needs to be perpetuated through a single process (and in most cases an automated process). This is where the proper cloud data management platform and strategy can accelerate cheap cloud object storage tiers as the endpoint for all backups and archives, regardless of the type of data, assuming that proper regulatory and compliance standards are met within your organization.
As a data administrator, regardless of where your application's data is stored, you want to be able to provide access to your data just like you always have. If you are Windows-based, you are most likely looking for an SMB/CIFS or iSCSI connectivity option to access your data. In the same sense, if you are Linux-based, you would be looking for connectivity to your data via NFS. If we are talking about advanced analytics or machine-learning processes, then you would want your data to sit in an object tier (such as S3). From the perspective of the application owner, how accessibility is granted is usually not at the forefront. However, to the storage or data administrator, you will want to maintain the same management through a cloud data management platform as you would if the data were being stored on-premises.
Depending on your organization, major challenges include compliance and governance responsibility across service models (IaaS, PaaS, SaaS) and how your organization ensures compliance with data either on-premises or in the cloud. The cloud data management platform should have built-in tools to help you manage this spectrum for your organization. The ability to scan data prior to a movement to the public, against various compliance standards (PCI, HIPAA, GDPR, etc.), is critical to ensure that only compliant data is being moved into a public cloud storage location.
There are several challenges with security and encryption as you are moving data out of your traditional data center. This can be access control policies, firewall and security group rules, role-based access controls (RBAC), user management and audit logging, and public vs. private network access. Plus, you add in the encryption components, hardware-level, software-level, data-at-rest, and in-flight, all while managing your security posture both on-premises and in the public cloud. With all the components above, you can see how maintaining security can be difficult within your organization. It is easy to see why a proper cloud data management platform with built-in security and encryption controls is extremely important as you consider moving data to the cloud.
Cost can be a critical factor in deciding on data placement and data needs to be stored as efficiently as possible. You will always incur costs for usage and ingress/egress of data traversing public cloud providers, managing data movement, storage efficiencies (thin provisioning, compression, deduplication), and its storage tier in the cloud, which makes these aspects key points to think about in your decision-making process.
Here are important things to look for when choosing a cloud data management solution:
- How does the storage platform handle storage efficiencies?
- How are snapshots managed?
- How does data cloning affect my usage?
- Is there a way to measure inactive (cold) data?
- Is there a built-in way to tier or move data to cheaper storage options, such as object storage?
- What are the data transfer costs, both inbound and outbound or across regions?
- What cloud resources are needed in order to deploy my storage management software in the cloud?
- What does a minimum viable product (MVP) in the cloud cost?
- What is the total cost of ownership of data as it moves to the cloud?
With the push for containerized-stateful applications and the use of Kubernetes for orchestration, you still need data to be persistently available to these applications regardless of where your data or application lives. The ability to manage Kubernetes persistent volumes within your cloud data management platform is key to effective management of your application data. If you can then take this one step further, from a data administration standpoint, and automate the request for data from the application (essentially putting control in the application owners' hands for how and when they need data), then you are then in a position to properly manage data for containerized applications without adding tons of manual work to your internal process.
Arguably the most important component of your cloud data management platform, but sometimes one that is overlooked. As you move more and more data from your traditional data center to the public cloud or multiple clouds, you need to be able to manage the environment without adding more human resources. The cloud data management platform should provide an avenue to end-to-end visibility, unified management, interoperability, and consistent security and compliance policies.
The cloud data management platform and strategy is not a simple direction, but a vital part of any organization's best practices. With many factors to consider as you manage your data, it is very important that nothing is overlooked. Based on the number of aspects to consider above, WWT can assist any organization to ensure how data should be managed, whether data is on-premises, only in the public cloud, or both.
To explore your cloud data management options further, contact us to get started.