Visualization on MPPA chip based Clusters and Smart Mobile Devices
This product line is a cluster of MPPA chips acting as an accelerator board. The PCIExpress boards containing MPPA chips are connected using multi Gbps Ethernet switch. Cluster configurations are customized by users. The software for the cluster is also customized by users. The system will be built based on customer's order; selected software will be installed and tested using specified applications before shipping.
Visualization on GPU Clusters and Smart Mobile Devices
We are developing visualization systems that are capable of generating interactive visualizations of enormous datasets that result from physics based simulations on High Performance Computing (HPC) facilities. The visualization system will support both end user applications and a toolkit which enables the development of custom visualization tools. Using scalable rendering techniques, the system will operate on the datasets as they reside on the HPC platforms and display the results on the display units. Our target market is bio-science research labs., pharmaceutical companies engaged in drug discovery and molecular therapy, multimedia authoring companies using high definition clips, and companies doing CFD and CEM.
This product line is a cluster of GPUs based on Dual Xeon processors with state of the art Graphics Processing Unit (GPU) and optional accelerator boards. The PCIExpress based GPU nodes are connected using a one Gbps Ethernet switch. Cluster configurations are customized by users. The software for the cluster is also customized by users. The system will be built based on customers order; selected software will be installed and tested using specified applications before shipping.
This product line is also a cluster of GPUs based on Dual Xeon processors but the software is tailored towards High Definition (HD) capturing, editing, and authoring. The PCIExpress based GPU nodes are connected using a one Gbps Ethernet switch. Cluster configurations are customized by users. The software for the cluster is also customized by users. The system will be built based on customers order, software installed, and tested using selected application before shipping.
The simulations in science and engineering regularly generate terabytes of data that needs to be analyzed via scientific visualization techniques. Moving that data to a remote location for visualization is impractical if not impossible. While techniques such as feature extraction and sub-sampling can be useful, it is commonly necessary for important visualization algorithms to operate on the entire dataset. Performing these algorithms on the HPC platform itself is a natural step in this process. However, during the past decade, the increase in the power of commodity graphics processing unit (GPU) cards has outpaced the native, integrated graphics sub-systems available on HPC platforms. Groups of inexpensive commodity GPU systems are used to construct multi-display systems. By combining the output of multiple displays via rear-projection screens, scalable display resolutions have been achieved. The challenge in the next decade is to make visualization possible on smart mobile devices. Smart phones already have support for 3D graphics and games. Visualization of scientific and simulation data on smart mobile devices will open up new applications in science, medicine, factory automation, and patient care. The Vizag line of products are aimed at this market.
We are also developing systems that could visualize structured and unstructured grid datasets using polygon-based and volume visualization techniques using HPC platforms for the algorithms and single or multiple smart mobile devices (SMD) for the display and interaction.
Structured grid dataset size can reach the multi-billion cell level while unstructured grid datasets are typically an order of magnitude smaller. Displays can be a single SMD or a multi-display co-planar system (DisplayWalls). The system would need a graphical user interface as well as a scripting interface for custom visualizations. While a single visualization application may be sufficient for many simulations, there will always be a need for customization. Therefore, the system should be capable of being utilized from a SMD using scripting languages. This will allow both the production visualization environment and the customizable tools to be based on the same subsystems.
Integrating computing, communication, and consumer electronics functions in smartphones is the focus of this project. Integrating novel low-power MPPA chips for supporting high end applications in scientific simulation, visualization, and medical image analysis is one of the goals. Stream based programming model will be used. A discrete-time block diagram based model descriptions will be used in the project to describe applications.
High performance and low-power MPPA chips that facilitates the use of a cellphone as a mobile, computing, communicating, and consumer electronics convergence (MC4) device is under development. The key idea is to provide a computing card similar to a SD card that can be added to a smartphone to support many functions. This approach allows the smartphone to be turned into a compute engine for multimedia processing, navigation processing, and control processing in a self organizing wireless web. Self organizing and self adapting software and protocols can be supported by using the proposed architecture in MC4 device to enable seamless connectivity for multimedia data. By supporting multiple radio bands in the basic cell phone network processing and wireless web communication over local, rural, and metropolitan areas could be done without investments in infrastructures.
The computing card could revolutionize the way consumer electronics items are designed. Medium to high quality digital cameras, video cameras, and music players have a high demand on computing. The proposed computing card could be used with these devices also to provide high quality results at a low-cost. The system architecture of the computing card based on a stream programming model and its communication with other parts of the smartphone are under development.
Universal telecommunication oriented personal information access (UTOPIA) project is aimed at rural e-commerce and seamless connectivity using cell phones. More than 70% of the world's population lives in rural areas and most of them are in Asia and Africa. Although rural e-commerce is touted to be the lifeline for rural Australia, rural New Zealand, rural parts of California, most of the rest of the rural world is yet to benefit from it. The reasons are inadequate roads, communication facilities, computer hardware, and the expertise needed to use and maintain the computer facilities. Other reasons are cultural, illiteracy, and price. Although infrastructure building is taking place in the rural areas of the world with help from IMF and WTO, a fresh approach is needed to bring the fruits of e-commerce to rural areas of the developing countries. The growth of cell phones and smart phones offers new opportunities for low-cost access to the World market from rural areas. Simpler user interfaces involving text and spoken commands, touch displays, and visual gestures are needed. Robust and reliable thin clients and cell phones are expected to be the user access devices for rural areas in the coming decade.
The project is developing approaches that could be used in realizing rural e-commerce.
The architecture of the village/town area network structure, sensor networks, and applications that are needed to do word/speech recognition, object recognition using cameras, and location gathering are under development. The impact of the portable device and the associated technology in conducting e-commerce from villages and remote areas in various parts of India, Korea, and Nigeria are planned.
Personal information access, news, alerts and warnings from many places and at different times is becoming a required activity in the current society. The spread of smartphones and mobile devices with sensors such as cameras, 3-D accelerometers, GPS, RFID receiver, and odor detectors with a computing card can open up new avenues for Information and personal access for all citizens of the world. While developed countries have access to information using internet connected PCs, laptops, and mobile devices, the timeliness of the information, ready access, securing the information, and sharing are becoming important issues. Seamless connectivity for mobile devices over WLAN, WMAN, and WAN is an open problem. Cost of accessing the information and convenience are also becoming major factors. Making personal information access a universal utility for the world has many economic, political, and cultural advantages and this is the focus of the ubiquitous computing, pervasive computing, and IT projects. Major Asian countries such as China, Korea, and India are planning on leveraging UTOPIA like approach to achieve and expedite economic development. Countries in northeast Asia (NEA) have a lot to gain by providing low-cost and flexible hardware, business models, infrastructure development, and project implementations.
CAGE is a methodology for architecture composition and mapping to scalable computing fabric (SCF) such as MPPAs. It is under development and some preliminary results are available. CAGE comprises four parts: a compiler front-end that produces IR code for programming languages, an analyzer that produces a skeletal architecture description from the IR code, an architecture generator, and an elaborator that produces a detailed architecture model at the component level or RT-level. The architecture model can be mapped to SCF such as MPPAs in one of two ways. The first approach assumes that a library of parameterized architectural components is available. Using the library of components an implementation of the architecture is assembled by interconnecting the components with the help of a router. The second approach uses core libraries, and partitioning, placement, and routing (PPR) tools provided by the MPPA vendors to implement the architecture model.
The CAGE methodology is being developed for video and vision applications. There are many applications that have been developed over the past four decades along with libraries. We want to be able to run these applications that contain parallelism on MPPA chip based systems. This approach complements the Simulink based (discrete-time based block diagrams) approach developed by the BEE/BEE2 group at UC Berkeley for high end signal processing in wireless and radio astronomy applications. Starbridge's Viva is a software environment that allows users to describe their algorithms using Viva's graphical development environment and maps it to FPGAs resources for direct execution. It is in some ways similar to the BEE approach except that Simulink is replaced by Viva's graphical language.