GPUs and flash memory are used in numerous defense and intelligence operations today and the number is rapidly growing. The need to store and transcribe huge amounts of data through data and image processing is becoming overwhelming. The more GPUs and flash memory available, the quicker the data can be used. GPU appliances supporting multiple NVIDIA GPUs and Flash memory appliances supporting multiple NVMe flash memory cards are quickly becoming the best and most economical way of accomplishing this tremendous feat.
Geospatial visualization (geovisualization) applications, used by the military to create real time mapping of the battlefield, require high compute acceleration to provide necessary data quickly. Traditional static maps are helpful for getting around but on the battlefield, a map that shows real-time data is much more valuable for ground troops. The Blue Devil 2 program is an example of how geospatial visualization would help the military. Cameras in a blimp gather data, computers analyze the data quickly and then send it to troops on the ground. Today these calculations are performed with specialized software such as Eternix Blaze Terra or GeoWeb3d running on GPU cards, coprocessors, or FPGA cards. The military gathers vast amounts of information from a variety of sources that needs to be manipulated to generate the 2D and 3D mapping required by field operations. GPU cards, with thousands of cores each, offload the number crunching and image processing from the CPUs.
On the battlefield, being able to know where everyone else is located, in real-time, is highly valuable. The Raytheon Company won a DARPA competition using a One Stop Systems FSA product that proposed a solution to this challenge of knowing immediately where every ground troop, humvee, aircraft and enemy combatant is at. In addition to processing large amounts of data quickly, geospatial visualization requires large amounts of storage for the tremendous amount of data that needs to be analyzed and stored. But a large amount of storage would not be helpful in this application if it weren't also incredibly fast. That's why flash memory is the best technology for the job. Not only can you store large amounts of data on a flash memory card, but you can access and save that data very quickly. When it's a matter of life and death, faster storage can save lives. One of Sandisk's Atomic Series ioMemory cards has a capacity of 6.4TB. The One Stop Systems Flash Storage Array puts 32 of these cards in one system that can be easily attached to any server running geospatial visualization software.
One Stop Systems, in conjunction with the U.S. Army, is providing Flash Storage for the DCGS-A Intelligence program. "The 'Distributed Common Ground System – Army' is the Army’s intelligence component that gathers intelligence data spanning all echelons from space to mud. DCGS-A enables decision makers to save lives immediately by gathering, analyzing and sharing intelligence information into a common system. This enterprise utilizes intelligence information and open source technology (such as Amazon, Google and Twitter) to create a scalable environment for collaboration and intelligence production. Overall, it has more than 600 sources of information! The enterprise has seven main capabilities: Planning and Direction, Collection Processing/Exploitation, Analysis, Prediction and Production, Battlespace Awareness Data Dissemination, and Relay. A few examples of the tools used under these seven categories are: analyze imagery or map products, process collected cell phone data, reporting human intelligence, matching biometrics, use NSA or aerial ISR Platform Data, and review full motion video. DCGS-A is more than just a work station for an intelligence analyst, or a software application. It is a global enterprise of capabilities that have allowed Soldiers to Succeed and Survive in two wars, and in fields across the globe to support world-wide intelligence operations." Source: https://dcgsa.army.mil/
Synthetic-aperture array radar (SAAR) is a form of radar that is used to create images of landscapes and other objects. The SAAR antenna is usually mounted on a moving aircraft or spacecraft that flies over the target region and uses the motion of the antenna to provide very detailed spatial resolution. The resolution achieved with SAAR is far more than is possible with traditional beam-scanning radars.
The Air Force Research Laboratory (AFRL) employs the OSS Accelerator because it can accommodate demanding signal processing applications, such as those with high bandwidth input, high computational requirements, and high bandwidth output. Real-time image and radar processing (such as SAAR) are examples of this type of processing applications. The High Density Compute accelerator can hold 16 NVIDIA GPUs, so it enables the high performance signal processing chain to achieve much higher performance than it would get without accelerators.
In addition to requiring a lot of compute power, the higher resolution data required to successfully map with SAAR also requires huge amounts of storage capability. The University of Dayton Research Institute utilizes the One Stop Systems 3U Flash storage array with Fusion-io ioMemory cards for their SAAR application. Anytime huge amounts of lightning fast storage are required, our product can provide the storage in a very small hardware footprint.
Designed to secure the Navy’s future in long-range maritime patrol capability, the P-8A has transformed how the force will man, train, operate and deploy. The P-8A operates with a smaller crew, yet it also delivers an extended global reach, greater payload capacity, higher operating altitude, open-systems architecture and significant growth potential.
The major sensor systems include an APY-10 radar system, which was developed specifically for the Poseidon and features inverse synthetic aperture radar, and synthetic aperture radar. Additionally, the Poseidon carries an electro-optical/infrared sensor turret and has increased acoustic capability to conduct concurrent passive and active processing. The P-8A Poseidon is responsive and interoperable with traditional manned forces and evolving unmanned sensors.
OSI Radar simulation and training application require lots of parallel compute power. This power is needed to quickly develop a suite of real-tim injection capabilities into the signal/data processor to demonstrate and test the sensor directorate's family of radars. Maximizing the number of CPU cores and GPU's is also extremely important. The more GPU's and CPU cores available, the quicker data can be used. OSS GPU appliances supporting multiple NVIDIA GPU's connected to a 2 - 4 socket servers create an ideal high performance compute acceleration platform used in algorithm development, data logging, analysis pre-flight mission testing, post-flight reconstruction, and training models. Freed up PCIe slots in the server are also able to support additional key IO such as GPS Synchronization, Timing, and Serial FPDP devices.
OSS provides a ful turn-key solution using scalable COTS technology designed to meet the increasing performance thresholds for today's requirements. Additional performance can be added through low risk, low cost upgrades driving a path to exceeding the objectives of tomorrow.
Here are some defense and intelligence applications that can be sped up by using GPUs:
Our compute accelerators support from one to sixteen double-wide PCIe cards and can be cabled up to four host computers through PCIe x16 Gen3 connections each operating at 128Gb/s. The all-steel construction chassis house power supplies, fans, and a system monitor that monitors the fans, temperature sensors and power voltages. Front panel LEDs signal minor, major or critical alarms. The compute accelerators are transparent and do not require software except for the drivers required by the PCIe add-in cards. Compute accelerators are the best appliance for applications that require a large amount of compute power.Learn More
Our flash storage arrays support from one to thirty-two single-wide PCIe cards and can be cabled up to four host computers through PCIe x16 Gen3 connections each operating at 128Gb/s. The light-weight, compact chassis house power supplies, fans, and a system monitor that monitors the fans, temperature sensors and power voltages. Flash storage arrays are the best appliance for applications that require a large amount of fast, flexible storage.Learn More