Big Data needs storage software defined – CIO

The proliferation of mobile devices and instrumented corporate assets is igniting a new explosion of data, from which can be acquired new analytical perspectives and, in turn, open up new business opportunities. At the same time, big data is to place a greater demand on existing infrastructure, driving the need for immediate access to resources – computing, storage and networking – and the creation of a new imperative to adopt cloud technologies. The flexibility simply can not be achieved with an approach focused on traditional hardware.

Storage is a constant pain point for cloud deployments, but has been largely ignored by IT organizations that have focused their attention primarily on server virtualization and network. With the growth of capacity, application’s performance, and issues related to cloud organizations that challenge, IT managers need to improve storage efficiency not only virtualizing your server infrastructure, but also their environments storage.

according to a study conducted in 2013 by the EMEA study, and provisioning storage management is an important bottleneck to 58 percent of the company cloud implementations. As a result, storage automation was identified as the demand for higher integration for the initial launch of cloud projects, as cited by 32 percent of organizations.

For the respondents who had tried to deploy a cloud private, without an infrastructure SDS (storage software defined), an 84 percent landslide of them were planning some sort of hardware independent storage virtualization system to support your cloud.

by abstracting service underlying proprietary hardware store by SDS can improve operational efficiency, provide transparent data mobility and enable common denominator management capabilities, regardless of the hardware used. SDS makes applications more efficient cloud, reducing management complexity and enabling them to scale cheap. A review of Worldwide Software Based (defined by the Software) IDC Taxonomy Storage, 2013 [1] report, together with a consensus of storage vendors indicates SDS has the following key attributes:

• 1 – Software is at the heart of the SDS. It was designed to run on heterogeneous commodity hardware and can even leverage existing storage infrastructure of an organization.
• 2 – SDS provides a complete set of storage services.
• 3 – SDS federates physical storage capacity from multiple locations, and internal drives, flash systems, other external storage systems, then from the cloud platforms and cloud objects.
• 4 – SDS is easily programmable via a single API, unified, which is available through a variety of portals

With reduced complexity, SDS reduces the load for administration. Simplifies, virtualizes and automates storage services. It improves utilization of existing storage and facilitates the migration of data between storage systems and storage levels – even among suppliers. – Thus reducing the delivery time of services

Here are some features that are important to have in any SDS infrastructure:

SDS should be opened. SDS should support wide range of customers in the physical storage infrastructure and integrate with other virtual computing and cloud management software. This openness helps organizations migrate to an agile storage environment, cloud-based and manage it effectively, without having to replace existing storage systems, generating dramatically increased value of existing investments. It offers virtualization-based storage capacity and automation, enabling customers to deploy the necessary resources without licensing complexity.

SDS should be self-optimization, intelligent, policy-driven. SDS should automatically adapt to workload changes to optimize application performance by eliminating most manual tuning efforts. Automated tiering across storage systems and suppliers of virtual machines and brands can optimize storage, moving the most active data automatically to the fastest storage level.

SDS need to be aware application. SDS should automate provisioning and increase productivity for administrators of IT operations can focus on the overall deployment storage and use, as well as the strategic requirements of long-term – without being distracted by routine storage provisioning requests. For example, the right storage solution would make instant backups frequency recognition applications throughout the day to reduce the risk of data loss.

The value of SDS in the real world
consider this example cloud and virtual infrastructure: a financial services company is starting to build a private cloud infrastructure defined by software. Has your virtualized infrastructure and computing is enabling network and storage virtualization in the next phase. Its essential business applications for processing credit card and virtual desktop term in a Dallas data center and a large data credit risk application, along with an application development cloud is running in all centers . data Beijing and Sao Paulo

The credit risk application for this type of organization is a big data application that must process data in real time from multiple sources: credit card transaction data, data bureau credit score, personal customer data, Twitter feeds, and other social media data publicly available. The intensive credit risk data application running in the cloud Beijing and begins to experience an I / O SDS A good infrastructure detects the subject and through a policy change to the storage layer, automatically new flash storage provisions. The SDS system then automatically changes the data “hot” relevant to this new flash storage, improving I / O, and maintain service-level agreement.

With SDS, the data can be moved from common dynamic and continuously, the storage capacity can scale elastically, and new levels of performance can be made in a transparent manner. In our example, the credit risk application invaded the physical infrastructure that had been assigned. The SDS system responded, automatic provisioning of new resources and move data as needed

When SDS is well done, the main benefits emerge:.

• 1 – SDS automates the use of on-premise and on-cloud storage resources
• 2 – Orchestration based storage resource policies optimizes performance and efficiency
• 3 – Analytics- Optimization driven software-defined resources can meet unpredictable business needs
• 4 – Based on open APIs, tools and technologies maximize customer value, availability of skills and easy reuse in hybrid cloud environments

vendors are already competing to claim the SDS space, each with its own approach. For example, IBM Virtual Storage Center allows automated, policy-based storage tiering and virtualization of heterogeneous storage systems and can existing storage turn into private cloud storage, no “rip and replace.”

a new vital component of the software-defined data center, SDS is still evolving – and will continue to do so at a rapid pace. According to an IDC 2013 report “storage-based software will slowly but surely become a dominant part of each data center or as a component of a data center defined by software, or simply as a medium to store data more efficiently and cost-effectively compared to traditional storage. “

the SDS right solution is one that helps customers transition from traditional storage infrastructure to an environment defined by agile software ready for cloud and manage it effectively. Deployed properly, simplifies and modernizes SDS heterogeneous storage environments using data management-driven analytics to reduce the cost of storage, and standardizes advanced data protection storage systems.