How to Design Self-Service Storage – ioFABRIC Briefing Note

For application owners, storage is an overly complicated ordeal that they must go through in order to get applications up and performing well. Ideally, application owners should be masked from the complexity of the infrastructure. They shouldn’t care which storage system their application is running on or what method of data protection is used to protect it. The problem is that IT needs to care. Storage systems have varying strengths and weaknesses. IT needs to mask this complexity from users.

The Self-Service Front End

The front end of a self-service storage infrastructure should present the user with a portal that asks basics service level questions like: How much capacity does an application need? How much performance and what level of protection does the application require? Simple dials or slider bars will give the user all the granularity they need. After these knobs are turned, the back-end infrastructure should automatically respond and present the application owner with storage that meets their specifications. It should also adapt over time to continue to make sure that these requirements are met or alert the storage administrator if they can not.

The Self-Service Back End

The back-end of the infrastructure needs to deal with reality. It needs to be part discovery tool, part orchestration tool and part monitoring tool. Most data centers have and need multiple storage systems to meet user demands. A mixture of all-flash arrays, hard disk arrays and network attached storage (NAS) systems are likely present in the data center. All should be available to the self-service back end. In addition, the back end infrastructure should also leverage in-server storage resources like internal flash drives, hard drives and even DRAM.


ioFABRIC is a software solution that provides a self-service front end for application owners as well as a discovery, orchestration and monitoring solution for the back-end infrastructure. In the ioFABRIC case, the application owner creates storage based on the application’s requirements using slider bars that affect performance (latency, IOPS and throughput), capacity and protection (snapshots, replicas and encryption).

The discovery component continuously audits the IT storage infrastructure to understand and organize what assets are at its disposal both initially and as new resources are added. The resources are then organized by capability (performance, capacity, location, etc..).

When an application owner makes a request for a certain type of storage, the orchestration capability creates the “best match” based on available resources. Finally, the monitoring tool continuously audits the environment to make sure the system is meeting service levels. If it is not, then the software solution communicates with the orchestration tool to transparently move data to return performance to agreed to service levels. Transparency is critical, the storage administrator nor the application team have time to update applications to look for data on different storage systems.

ioFABRIC Architecture

The ioFABRIC data plane is a virtual volume presented as a virtual device. That virtual volume is made from the various physical storage resources IT makes available to it. It can consist of both internal server storage, on-premises shared storage and cloud-based storage. The storage media can range from DRAM/NVDIMMs to flash to hard disk drives. ioFABRIC can even pool available DRAM from multiple servers into a volume, creating an extremely high performance tier for the most demanding applications.

The ioFABRIC Control Plane manages performance and capacity as well as ensures it meets data protection objectives. IT interacts with the control plane via ioFABRIC’s web-based dashboard or a command line interface, both will integrate to a REST API. Larger organizations can bring storage provisioning and monitoring into enterprise orchestration tools via this REST interface.

The solution provides a full complement of data services to the storage it manages, including snapshots and replication. The solution features multi-site high availability to make sure its services continuously run.

Part of ioFABRIC’s reach is into the cloud. It can move data to the cloud based on policy. Once in the cloud, it can control data residency making sure regulated data starts and stays within country. It can also compress, deduplicate and encrypt cloud-based data to reduce costs and increase security.

StorageSwiss Take

Self-service is an IT buzzword that is on every CIO’s project whiteboard. But creating a self-service environment is challenging, especially for something with as many disparate parts as storage. ioFABRIC is doing an admirable job of not only creating a fabric backend for storage but also creating a very usable front end for application owners. CIOs looking to get self-service storage off the whiteboard and into production should give ioFABRIC a close look.

Twelve years ago George Crump founded Storage Switzerland with one simple goal; to educate IT professionals about all aspects of data center storage. He is the primary contributor to Storage Switzerland and is a heavily sought after public speaker. With over 25 years of experience designing storage solutions for data centers across the US, he has seen the birth of such technologies as RAID, NAS and SAN, Virtualization, Cloud and Enterprise Flash. Prior to founding Storage Switzerland he was CTO at one of the nation's largest storage integrators where he was in charge of technology testing, integration and product selection.

Tagged with: , , , , , , , , , , ,
Posted in Briefing Note

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 22,131 other followers

Blog Stats
%d bloggers like this: