Software Advantage for Storage Virtualization
For DataCore
This is a Press Release edited by StorageNewsletter.com on January 3, 2011 at 3:59 pmDataCore Software highlighted what it considers points of emphasis for 2011 in terms of furthering the promise of virtualization.
The promise of virtualization will be further realized this year. Software-based storage virtualization will take its rightful place by making storage anonymous within virtual infrastructures.
A number of trends, strategic drivers and software architectures will emerge to make a new level of flexibility and scale both practical and cost-effective to achieve. The most significant driver will be the simple realization that only a software-based infrastructure can deliver the advantages to make virtualization and clouds a reality because software can abstract itself from device specific limitations, provide portability across platforms, and endure beyond the life of underlying hardware platforms that come and go over time.
Compelling Benefits:
The Software Advantage and Anonymous Storage
According to DataCore Chairman and Co-founder Ziya Aral, "The entire storage virtualization angle was real simple – abstract yourself from the hardware. DataCore was founded on the belief that storage controllers in general and storage virtualization in particular were essentially ‘software programs.’ For us, virtualization was driven by a very simple need to make a portable software program to do disk storage and run it on any platform."
This sentiment is both echoed and emphasized by DataCore’s President, CEO and co-founder George Teixeira, "From a business standpoint, the total cost of ownership and the payback on investment of a pure ‘software infrastructure,’ where you pay once for intelligent software to manage, protect and get more from your storage assets – as they come and go from generation to generation and brand to brand – is a value proposition that is as compelling as it is inevitable."
Bottom-line: A software-based storage virtualization infrastructure can live within the same abstraction level as virtual servers and virtual desktops and bring along the same type of benefits that we have seen from these movements in regards to making storage hardware brands largely irrelevant – or anonymous – to users and applications, therefore simplifying management, removing storage task complexity, speeding up response and provisioning times and increasing overall utilization and flexibility.
The ‘software advantage’ becomes obvious when you consider the many scenarios it works in and the many shapes it takes and will take in the future. Because DataCore software is portable, it can run on a virtual machine (VM) or on a multitude of physical servers. It not only virtualizes and manages storage, but can coexist along with the server hypervisor in the virtualization layer, providing many possibilities to solve real world and real budget challenges. Hardware can’t do that.
Software-based storage virtualization has been lauded as a ‘game changer’ in a number of reports, including: InformationWeek‘s ‘Storage Anonymous’ Cover Story.
DataCore’s Future Directions
for Storage Virtualization
"The whole idea of DataCore and our vision was that if you could move the software program from the constraints and boundaries of the physical platforms and take it to a modern programming environment, then it would be possible to do with software what had been done manually to that point – that’s storage virtualization," explains Aral.
Virtual desktops (vDesktops) are an example that highlights the need for a new model for storage. The major challenge for vDesktops is that SANs are often implemented with large and costly storage controllers and complex external storage networks. While these have the advantage of achieving reasonable scalability, they introduce a large threshold cost to virtual desktop implementations. To overcome this high capital cost burden, hardware vendors typically tout the economics of deploying several thousand vDesktops. The real problem in this case is not in scaling up to thousands of vDesktops but in scaling them down to practical configurations. It barely needs mentioning that this must occur without radically spiking costs at the low end and also without forgoing the SAN feature set which assures portability, availability, and data redundancy. Otherwise, the very benefits of vDesktops are compromised.
DataCore has done benchmarking to understand the economics of vDesktops and has been able to configure high-availability configurations supporting a few hundred instead of thousands of vDesktops, and at a cost per desktop of less than one tenth of what was previously reported. In addition, new topologies allow the scaling of such configurations to thousands if that is what is required. Based on these initial findings, DataCore will make an impact on removing storage costs as a primary barrier to virtual desktop deployments. DataCore will be posting a series of Virtual Desktops Benchmark Reports in 2011.
Virtual desktops have one element in common with the other major computing movement of our day: cloud computing. Both technologies promise to deploy very large numbers of machines of the same class. This creates the opportunity to present one additional level of virtualization and to ‘divide and conquer’ the problem which is otherwise daunting for its scale. What if, instead of attempting to manage hundreds or thousands of virtual machines discretely, one could divide them into arbitrary groups or sub-units and then manage a far smaller number of sub-units? This will also be the future direction of our work, impacting not only vDesktops and cloud computing but also the organization of storage itself.
Comments