Market Overview and Definition

Software-Defined Storage (SDS) provides enterprise-class storage using commodity hardware and specialized software to deliver storage services, advanced features, and management capabilities. This differs from the traditional enterprise storage approach requiring proprietary or custom storage systems with higher up-front acquisition and on-going operational costs.

There is growing confusion in the market about this term since some vendors are trying to use the term software-defined storage to define storage management software (“Control Plane” only). For this overview, Maxta is referring to the purest definition of software-defined storage that includes both “Control Plane” and “Data Plane”. SDS is sometimes also referred to as a Server SAN or as a hyperconverged or hyperconvergence infrastructure when delivered preinstalled on a server. Research analyst firm IDC states Virtual SAN is a high growth market with 41% year over year growth for the next two years1. And for storage markets, software based storage revenue is the fastest growing segment.

Software-defined storage is also part of a larger industry trend that is called software-defined data centers (SDDC). Moving storage intelligence to software facilitates the introduction of more advanced solutions into enterprise data centers, such as hyperconverged solutions that implement both compute and storage functionality on the same server for virtualized server infrastructure (VSI) and virtualized desktop infrastructure (VDI). SDS solutions are typically delivered either as software-based, reference architectures or may come pre-bundled in an appliance. Noting that when SDS is pre-bundled in an appliance some of the choice, and scaling flexibility is lost. Maxta and VMware VSAN are leading software-based solutions. Maxta with MaxDeploy and VMware with EVO:Rail are reference architectures that are known to successfully compete, based on their software-based flexibility, with appliance based hyperconverged solutions from Nutanix and Simplivity.

A software-based software-defined storage solution increases flexibility by enabling data centers to leverage servers from any vendor as well as to use existing infrastructure to create an enterprise storage solution. Additionally, data centers can achieve massive scale by increasing storage capacity and performance as needed, in any increment and without interruptions. The ability to run on any x86 server and any server-attached storage device prevents vendor lock-in and provides flexibility of choice enabling significant CAPEX and OPEX savings. A reference architecture solution based on SDS, AKA a hyperconverged solution, is a preconfigured and pre-validated solution that comes with the additional benefits of simplified ordering and predictability of performance.

Evolution from Traditional IT to SDS

SDS solutions are designed for limitless configuration flexibility and the freedom to choose from a myriad of compute, storage, and networking resources and provide enterprise class storage services, advanced features, and management capabilities. This differs from the traditional IT (hardware-defined) approach to enterprise data centers requiring proprietary or custom storage systems with higher up-front acquisition and on-going operational costs.

Enterprise data centers that have adopted cloud, virtual server infrastructure (VSI), or VDI face challenges due to the continued use of traditional enterprise storage systems. Popular virtualization platforms such as VMware vSphere, Microsoft Hyper-V, and Linux KVM that are the basic building blocks of cloud, VSI and VDI leverage software-defined technology at the compute level. However, traditional enterprise storage systems such as SAN (Storage Area Network) and NAS (Network Attached Storage) remain inseparable combinations of hardware, software, and services that can’t run on the same platform as the virtualization software. As a result, SDS that is built to run alongside virtualization on the same platform has become the most cloud- and virtualization-friendly storage solution.

 

Software-defined storage

Traditional storage

Hardware

Industry Standard

Proprietary

Software

Flexible

Proprietary

Configurations

Flexible

Pre-determined

Scalability

Scale up & Scale out

Scale up

Management

Integrated with Hypervisor

Separate from Hypervisor

 

To help lower costs and improve IT agility, vendors such as VCE, Cisco, Dell, HP and others, began providing converged solutions. These solutions include servers, data-storage devices, networking equipment and software for IT infrastructure management, automation and orchestration. A converged infrastructure addresses the problem of siloed architectures and IT sprawl by pooling and sharing IT resources. Rather than dedicating a set of resources to a particular computing technology, application or line of business, converged infrastructure creates a pool of virtualized server, storage and networking capacity that is shared by multiple applications and lines of business.

Software-defined storage takes this evolution one step further by providing the ability to run both compute and storage on any x86 server and utilize any server-attached storage device eliminating entirely the need for expensive and complex storage arrays thus enabling significant CAPEX and OPEX savings. Additionally, SDS solutions assimilate a global namespace and provide the functionality of shared storage such as data mobility, migration, load balancing, high availability and disaster recovery. VM-centric solutions also provide greater simplicity by eliminating the need for managing and provisioning LUNs and volumes, mapping VMs to volumes, and other tasks that are traditionally required in managing other storage solutions in a virtual environment.

The Main Drivers Enabling SDS Now

Since SDS solutions provide enterprise class reliability, availability, serviceability, and performance leveraging commodity server hardware platforms and commodity server-side storage, they introduce new opportunities that are not possible with traditional enterprise storage. Large Internet companies have been using SDS, albeit tailored for a specific use case, for quite some time to differentiate and lead in their respective area. For the same reasons, innovative data centers are now built using SDS solutions delivering enterprise storage at lower acquisition costs and operational expense.

OpenStack, an open source modular architecture for creating clouds and data centers, brings the promise of savings and elimination of vendor lock-in to IT. SDS solutions work well in an OpenStack environment. Solutions like MxSP that provide the entire suite of storage and data services enable hyperconvergence in an OpenStack environment, without compromises to enterprise-class capabilities.

Forecasts from leading industry analysts vary, due in part to differences among their unique SDS definitions, but there is broad consensus for rapid SDS growth versus traditional storage architectures. Today’s data centers are experiencing many significant benefits from software-defined storage including increased flexibility, automated management, cost efficiency, and limitless scalability.

SDS—The Killer Application for the Software-defined Data Center (SDDC)

Traditional enterprise data centers are expensive to manage since their inherent complexity requires a high-degree of specialized skills, knowledge, and abilities among IT staff. Enterprise data centers are full of traditional storage systems and designed to support a scale-up architecture. While expansion is possible to a point, scale-up storage requires expensive upgrades to support data center growth—eventually leading to replacements requiring a huge capital investment. All of this dramatically increases total cost of ownership (TCO) and reduces return on investment (ROI).

·      Traditional enterprise storage is a hardware-centric system.

·      SDS and SDDC are hardware-agnostic and software-defined solutions.

·      SDS matches and complements SDDC in ways traditional storage cannot.

SDS solutions like Maxta’s MxSP offer a pragmatic alternative to the inefficient replacement of traditional enterprise storage systems with larger or newer models. They offer the flexibility of scale-up and scale-out to support a broader range of data center workloads than traditional storage alternatives. Lower TCO is realized from lower capital expenses by using commodity x86 servers and server-side storage devices and from lower operating expenses from reduced costs in power and cooling. In addition, SDS solutions simplify IT management and are a natural complement to the software-defined data center.

Common Use Cases for SDS

Software-defined Storage (SDS) solutions like those from Maxta are deployed in a wide range of customer environments enabling various types of workloads including mission-critical applications, virtualized desktop infrastructure (VDI), remote offices and branch offices (ROBO), disaster recovery (DR), and test and development. Additionally, Maxta solutions are in use by industry leading cloud service providers as a foundation for their client services. SDS is the ideal choice for organizations moving towards a software-defined data center approach.

Criteria to Evaluate SDS Solutions

SDS complements other solutions within software-defined data center environments, delivers greater freedom of choice over storage configurations, lessens the importance of proprietary hardware, and increases the value of software. SDS solution criteria should leverage these benefits. Maxta recommends the following checklist of software-defined storage features, functions, and benefits as key criteria for evaluating potential SDS solutions for enterprise data centers, cloud infrastructures, and similar environments.

Choice

•        To use any x86 server whether it is a brand-name or a “white box”

•        To run on any server model up to the latest and greatest generation

•        To run on any hypervisor or abstraction layer (e.g., Docker Containers)

•        To use mixed drive types such as Flash, SSD, and spinning disk in any configuration

Manageability

•        VM-centric data services such as snapshots and clones

•        Single pane of glass for VM and data management

•        Pre-configured and pre-validated reference architectures

Scalability

•        Global name-space

•        Scale-out and Scale-up

•        Scale compute and storage independently

•        Flash optimized

•        Log based data layout

Resiliency/HA

•        Data availability, strong checksums for data integrity, and RAID support

•        Local mirroring and local replication

•        Metro-cluster support

Enterprise-Class Data Services

•        Highly efficient snapshots & clones

•        Ability to co-locate VM & associated data

•        Data Protection Policies: Schedule creation & retention of snapshots per data protection policies

Capacity Optimization

•        Inline compression and deduplication

•        Thin provisioning

•        Space reclamation

Cost Efficiency

•        Simplify management and provide VM-centric administration for OPEX savings

•        Ability to run on any x86 platform and utilize capacity optimization for CAPEX savings

Software-defined Solutions by Maxta

Maxta’s software solution, also known as MxSP “Software-Defined VM Storage Platform”, delivers enterprise and service provider customers with the greatest opportunity to modernize and simplify the management of their clouds and virtual data centers. MxSP provides customers the greatest choice of server platforms, hypervisor platforms, and storage configurations, all without compromise to required enterprise-class data services. MxSP is a resilient, scalable and distributed peer-to-peer architecture that aggregates storage resources from multiple servers. It assimilates a global namespace and provides the functionality of shared storage such as data mobility, migration, load balancing, high availability and disaster recovery. The MxSP VM-centric design provides simplified management and a single pane of glass management since it runs within the existing hypervisor management console. MaxDeploy reference architectures are also available to help accelerate adoption by providing partners and customers with the blueprints needed to deploy software-defined clouds with Maxta software and a large ecosystem of servers. Maxta MaxDeploy preconfigured and pre-validated solutions provide software-based reference architectures for multiple server vendors and infrastructure software vendors (e.g., OpenStack distributions). Solution briefs contain testing results providing predictability of performance based on configurations and workloads.

Unlike other companies that require you to purchase dedicated appliances with fixed configurations and expensive upgrades, Maxta’s solutions deliver the flexibility to use any server, any hypervisor, and any storage hardware that you want in addition to the ability to scale your compute and storage independently.

With MaxDeploy and MxSP solutions, Maxta customers have the utmost flexibility to customize their solutions, dramatically simplify their IT management, and to experience significant CAPEX savings in addition to OPEX savings.

Storage has not kept pace with compute in terms of the simplicity, availability, agility, and overall cost. The 451 Research findings show significant demand for SDS and hyperconvergence, including the need for the simplification of IT, enterprise-class data services, and cost efficiencies. By offering the choice of MxSP VM-centric software-defined storage and MaxDeploy family of hyperconverged solutions, customers have the maximum flexibility to choose the best solution for them with the complete flexibility to run on the server, hypervisor, and storage of their choice. – Yoram Novick, CEO, Maxta, Inc.