SBIR-STTR Award

This Small Business Innovation Research Phase I project will focus on developing a desktop-specific distributed architecture to provide enterprise grade, highly available and scalable virtual desktop systems that cost no more than a regular PC. In order to accomplish this, the focus of the research will be 1) to create a seamless, scalable and self-managed distributed fabric that is easy to deploy and 2) to develop an approach that will allow the distribution of large desktop templates (in the order of tens of gigabytes) to tens of thousands of desktops in less than an hour. This Small Business Innovation Research Phase I project takes advantage of a sizable and growing demand for virtual desktops that offer lower management costs and superior data security. However, existing solutions on the market many times the cost of regular PCs and require highly skilled personnel to operate, thus hampering their adoption. If the research team is able to overcome the technical challenges and create a solution that costs no more than a regular PC, it may be able to open up the market consisting of an installed base of 500 million commercial desktop PCs with 84 million new units purchased each year worldwide. This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5)

This Small Business Innovation Research (SBIR) Phase II project will focus on scaling and developing a desktop-specific distributed grid to provide enterprise grade, highly available virtual desktop systems that are cost-competitive with PCs. To accomplish this objective the research and development will focus on three key areas. First it will focus on multiple ways to scale the solution including ways to distribute large desktop templates across the grid, develop self-managing software to automatically update, patch and repair the software that manages grid and support a hierarchy of grids that can be geographically distributed. Second, it will develop multi-tenancy mechanisms to securely house multiple customers on a grid thus making this a viable solution for managing and economically providing desktops as a service from a cloud. Finally, additional work to support multiple hypervisors and remote protocols will be developed to address market requirements for cost-effective virtualization technologies and to provide good user experience when deploying the solution across a cloud. This takes advantage of a sizable and growing demand for virtual desktops that offer lower management costs and superior data security. Existing solutions on the market are 4 to 10 times the cost of regular PCs and require highly skilled personnel to operate, thus hampering their adoption. Kaviza's approach, if successful, reduces that costs drastically and makes it possible for the desktop IT staff who procure and manage PCs today, to be able to install and deploy proposed solution within their budgets. Kaviza's solution, if successfully deployed, also has the potential for broad economic and environmental impact. First, it can reduce PC power consumption by up to 75% making offices more "green". Second, its lower cost and simpler management will make it possible to broadly deploy computational resources and bridge the digital divide