COTS (kots), n. 1. Commercial off-the-shelf. Terminology popularized in 1994 within U.S. DoD by SECDEF Wm. Perry’s “Perry Memo” that changed military industry purchasing and design guidelines, making Mil-Specs acceptable only by waiver. COTS is generally defined for technology, goods and services as: a) using commercial business practices and specifications, b) not developed under government funding, c) offered for sale to the general market, d) still must meet the program ORD. 2. Commercial business practices include the accepted practice of customer-paid minor modification to standard COTS products to meet the customer’s unique requirements —Ant. When applied to the procurement of
for the U.S. Military, COTS is a procurement philosophy and does not imply commercial. Office environment or any other durability grade. E.g., rad-hard components designed and offered for sale to the general market are COTS if they were developed by the company and not under government funding.
By John Reardon, Editor-in-Chief, COTS Journal
Odile Ronat, SiTime, Director of Product Marketing
EDITORIAL
EDITORIAL SUBMISSION AND RELEASES
johnr@rtc-media.com
PUBLISHER/CEO
John Reardon • johnr@rtc-media.com
CONTRIBUTING EDITORS
John Reardon, Odile Ronat
ART AND PRODUCTION
CREATIVE DIRECTOR
Shae Reardon • shae.reardon@gmail.com
DIGITAL MARKETING MANAGER
Scott Blair • scott@blair-media.com
AD TRAFFIC
Vaughn Orchard • vaughno@rtc-media. com
FINANCE AND ACCOUNTING
Paige Reardon • paiger@rtc-media.com
ADMINISTRATION/CIRCULATION
Dawn Dunaway • dawnr@rtc-media.com
ACCOUNT MANAGER
SALES MANAGER
Vaughn Orchard • vaughno@rtc-media. com
CUSTOM AND CONTRACT SERVICES
John Reardon • johnr@rtc-media.com
(949) 226-2050
PUBLISHER’S NOTE
John Reardon, Editor-in-Chief, COTS Journal
Measured Response in a Millisecond World
The general stood at the edge of the observation deck, watching an attack unfold on the wall of monitors. Hypersonic blurs cut arcs across virtual skies, interceptors rose like reflexes from a sleeping giant, and data—events happening faster than thought.
And yet, beneath the hum of advanced tech, one question lingered like a distant storm cloud: Who decides to strike back when milliseconds matter?
As Publisher of COTS Journal, I’ve walked the corridors of innovation long enough to know the difference between progress and pretense. Artificial Intelligence—AI—is no longer a lab experiment. It is the silent observer, the always-listening sentinel. And in the theater of military defense, it is rapidly becoming both shield and sword.
But here’s the uncomfortable truth: Many of today’s loudest voices on AI in the military are missing the point. Some are enchanted by AI’s power to surveil, control, or even paci-
fy civilian populations—tools of oppression masquerading as innovation. Others focus solely on weaponization, envisioning autonomous machines waging wars beyond human reach.
That’s not the future I see.
For all its steel and fire, the military is, first and foremost, a human institution. It is governed by tradition, rank, and doctrine forged through generations. This hierarchy is not a relic—it is a safeguard. It ensures deliberation before escalation and control before chaos. And as battle plans grow ever more complex and threats arrive at speeds beyond comprehension, that chain of command must endure—even if its next link is artificial.
So, how do we introduce AI into this sacred framework without unraveling it?
The answer is focus. Narrow, intentional focus. AI must not be handed the launch keys in a moment of fear or aggression. Its role must be defensive by design, limited in scope, and rooted in one goal: to buy us time.
Consider this: A hypersonic missile traveling at Mach 5— more than a football field every 100 milliseconds. That’s faster than the blink of a human eye. Faster than most ethernet packets. In such a moment, no general can call a meeting. No commander can consult a doctrine. The decision must be made before anyone knows a decision needs to be made.
That’s where AI earns its place—not as a judge or executioner, but as an early responder. A buffer between detection and decision. A sentry that sees the shadow of a strike and raises a shield before it’s too late.
Historically, our Anti-Ballistic Missile (ABM) systems gave us minutes of warning by tracking the arcs of ICBMs across the sky. But the age of long-lead threats is over. Today’s dangers come from stealthy submarines and mobile platforms, from missiles that alter course in flight and cross continents in minutes. There is no arc to follow—only moments to act.
And yet, as we dream of a homeland defense system akin to Israel’s Iron Dome, reality sets in like a cold wind. The scale of such a system for the United States—landmass, population centers, geography—makes it nearly unimaginable. Alaska alone is 80 times the size of Israel. Should it be pro-
tected equally? Is Los Angeles more valuable than New York? The cost could reach into the trillions—before we account for the requirement that all components be American-made or open-standard compliant.
What we envision, then, is not a dome but a mesh—a flexible, evolving network of sensors, interceptors, and AI-driven response nodes that work together, not in place of, but alongside, human leadership.
But while we defend against physical strikes, another war rages in silence.
Cyber conflict is the invisible battlefield of the modern age. Our defense infrastructure spans multiple branches, generations of systems, and alliances that stretch across the globe. Every connection is a door. Every protocol has a potential weakness. Imagine every Bluetooth headset, every Wi-Fi link, and every firmware update in the field. Now multiply that vulnerability by a hundredfold. That’s what the Department of Defense faces each day.
In such a chaotic ecosystem, AI becomes more than helpful—it becomes necessary. Not to police or dominate but to monitor, interpret, and warn. Like a sentinel posted at every gate, its role is not to act but to alert. It buys the time we no longer have and identifies the threat no human could parse fast enough.
So, no, AI in the military should not be feared as an overlord nor worshipped as a savior. It is neither. It is a tool—powerful, precise, and dangerous if misused.
And like every tool in the arsenal, it must be wielded with discipline, restraint, and above all—human judgment.
Inside Track The
Kratos Demonstrates Fully Virtualized SATCOM for US Army in Separate LEO and MEO Projects
In two separate demos for the US Army, Kratos, along with partners, showed how satellites in Low-Earth Orbit (LEO) and Medium-Earth Orbit (MEO) could be utilized to support more agile and resilient multi-orbit, multi-mission capabilities.
In the first demonstration, Kratos and satellite operator SES showed how a flexible, fully virtualized satellite communications ground system could facilitate simultaneous communication pathways for resilient SATCOM.
The US Army’s Combat Capabilities Development Command demo successfully showed how a software-defined network architecture could seamlessly support satellites on a “make-before-break” mode over SES’s O3b MEO satellite network.
“Make-before-break” refers to the ability to transfer communication sessions while the user transverses coverage areas of different satellites—an essential capability for leveraging MEO and LEO constellations.
However, to leverage multi-orbit satellites, MILSATCOM networks require a more agile capability than current hardware-based ground systems. OpenSpace®, the software-defined satellite ground system developed by Kratos, demonstrated the agility needed for these future-oriented operations.
In the second demonstration, Kratos, Telesat Government Solutions, and Cobham Satcom showed how a flexible network architecture could dynamically support simultaneous communication pathways for LEObased SATCOM.
LEO constellations offer the strategic benefits of lower latency connectivity and redundancy.
“Every mission has different requirements for connectivity, which require maximum flexibility for leveraging multiple satellites, networks, and network elements,” said Chris Badgett, VP of Technology for Kratos. “Only a software-defined approach can provide the level of adaptability at
mission speed to maximize those available network resources.”
Kratos has worked with several satellite service operators over the past year to test and verify the functionality of OpenSpace as a gateway and edge platform at each of the major orbital belts, GEO, MEO, and LEO. In future conflicts, multi-orbit operations will be crucial to seamlessly sharing information amongst joint military branches and international partners.
Both projects were funded through the Network Cross-Functional Team (N-CFT) established by the Army Futures Command.
The
Inside Track
ARES Security and MatrixSpace Provision Powerful Radar Integrated with AVERT MPO to Create Decisive Tactical Advantage for Expeditionary Forces
ARES Security and MatrixSpace announce a technology partnership that combines the AI collaborative sensing capability of MatrixSpace Radar™ with ARES Security’s AVERT Mission Planning and Operations (MPO) software for autonomous systems.
The AVERT MPO software enables mission planning and control for squads of ground and aerial robotic systems from a single interface in a remote tactical operations center (TOC). The expendable low SWAP-C MatrixSpace Radar, which can be deployed on ground and aerial autonomous platforms, brings unparalleled adversary, vehicle, and sUAS detection capability into the AVERT MPO common operational picture. Through this collaboration, customers, in particular branches of the Department of Defense, are now equipped with a decisive tactical advantage across a range of security and defense applications.
The new capabilities will be featured at the Department of Defense’s Thunderstorm 24.4 in early November; an invitation-only field technology showcase focused on innovative solutions for expeditionary military operations.
“Battlespaces and secure sites must detect many objects accurately, moving rapidly in all weather conditions, day and night. Building MatrixSpace Radar into the ARES solution takes it to the next level in terms of managing and monitoring these secure locations. Enabling rapid counter UAS initiatives by bringing advanced technological solutions to the DoD is the key focus of our partnership,” stated Lori DeMatteis, CRO, MatrixSpace
Dramatically increased detect and track capabilities for various large-scale security situations.
MatrixSpace Radar offers robust situational awareness of airborne and groundbased objects, regardless of lighting and weather conditions. This facilitates highly accurate detection and counter-uncrewed aircraft system (CUAS) capabilities, beyond visual line of sight (BVLOS) flight for uncrewed, autonomous, and tethered aircraft, and overall general awareness and security.
AVERT MPO integrates robotic platforms, sensors, and personnel using the tactical assault kit (TAK) into a common operational picture, providing critical data
Anduril and Ultra Maritime Announce Exclusive Partnership on Groundbreaking Autonomous Ocean Sensing Capability
Anduril Industries and Ultra Maritime are proud to announce an exclusive partnership to develop a significant new capability in next-generation unmanned subsea sensing, making it possible for US and allied nations to enhance their submarine detection capabilities rapidly and inexpensively. By bringing together
The partnership will combine Ultra Maritime’s innovative Sea Spear deployable acoustic arrays and AI-enabled acoustic processing with Anduril’s undersea auton-
for informed decisions and rapid response. It is unique in its ability to automate robotic sentries using trusted AI, intelligent pathing, and 3D modeling/simulation and is capable of integrating with over 450 sensors through an advanced C2 platform, providing a decisive advantage during tactical operations.
Ben Eazzetta, CEO of ARES Security, commented: “We are impressed by MatrixSpace’s powerful radar, which delivers new levels of 4D outdoor surveillance, enhancing detection range, accuracy, and adaptability to adverse conditions. Paired with AVERT MPO’s AI-driven control and intelligent pathing, the solution enables precise mission planning, faster threat responses, and optimized patrols, creating a more robust security ecosystem for any security force protecting personnel and assets.”
omy capabilities, including Dive XL and Seabed Sentry, plus Anduril’s Lattice communications framework, providing a major advance in distributed, low-cost undersea surveillance.
As enemy submarines grow quieter and greater in number, the need for low-cost solutions to submarine detection grows.
Autonomous systems provide a clear path to low-cost, low-risk anti-submarine warfare (ASW) capability but have been hampered by inadequate technology to date.
Issues of range, endurance, sensitivity, communication, scale, and more have limited advances in autonomous or uncrewed
ASW. Anduril and Ultra Maritime have developed an innovative approach to enabling autonomous undersea ASW.
The concept uses Anduril’s Dive XL to autonomously deliver and deploy Anduril’s Seabed Sentry modular undersea payload system, which in turn hosts Ultra Maritime’s Sea Spear lightweight sensing array as a payload. Once deployed, Sea Spear pushes Sonar processing to the tactical edge using modern AI techniques. The system then uses acoustic communications and Anduril’s Lattice framework to provide users with real-time, autonomous submarine sensing. Multiple systems can be deployed and distributed, enabling users to see and detect signals distributed across large ocean areas. The joint effort is well underway and achieves end-to-end in-water testing in 2025.
Inside Track
Boulder Imaging has partnered with Roc Conservation Tech to expand IdentiFlight’s global reach and enhance wildlife conservation within the wind energy sector. With a shared commitment to bridging technology and environmental stewardship, Roc Conservation Tech will play a pivotal role in growing IdentiFlight’s presence and impact in South Africa’s renewable energy industry.
Inside Track
is embedded in square lattices of superconducting qubits and requires many physical qubits. Rigetti’s concept will leverage Quantum Low-Density Parity Check (qLDPC) codes, which do not require as high of a physical qubit overhead, making them a more efficient QEC technique for scaling toward fault-tolerant systems.
chip fabrication technique -- Alternating Bias Assisted Annealing (ABAA) -- that enables precise qubit frequency targeting and innovative multi-chip architecture technology. In recent years, Rigetti has deployed quantum systems that leverage vertical I/O, multiple layers of signal lines, and the ability to connect qubits on separate chips with high fidelity.
On the brink of a renewable energy revolution in South Africa, Roc Conservation Tech is uniquely positioned to assist as the country enters this transformational era by specializing in leveraging data science and engineering for conservation, with expertise in bird detection systems, conservation drones, and environmental monitoring. The collaboration between the two companies will promote the IdentiFlight mission to help mitigate the cumulative impacts of wind energy on biodiversity. The IdentiFlight system operates with 99 percent accuracy and reduces bird fatalities in wind farms by more than 85 percent in 12 countries across five continents.
“The need for sustainable energy solutions comes with the responsibility to protect biodiversity,” said Dirk van der Walt, founder of Roc Conservation Tech. “South Africa’s wind energy sector is rapidly growing, yet it is home to diverse and vulnerable bird populations that require advanced conservation measures. Partnering with IdentiFlight to deploy these systems in South African wind farms will help ensure that renewable energy expansion in the region is both responsible and sustainable.”
“Strong partnerships are essential to Boulder Imaging’s mission and IdentiFlight’s continued growth,” said Don Mills, president and chief operating officer of Boulder Imaging. “By collaborating with Roc Conservation Tech, we are bringing a wealth of local expertise and resources to the South African market, enhancing our ability to efficiently and responsively serve our customers.”
Computing Selected to Participate in DARPA’s Quantum Benchmarking Initiative
DARPA’s QBI program objective is to verify and validate that quantum computing can achieve utility-scale operation by 2033. Rigetti Computing, Inc., a pioneer in full-stack quantum-classical computing, announced today that it was selected to participate in the Defense Advanced Research Projects Agency (DARPA) Quantum Benchmarking Initiative (QBI). The primary goal of QBI is to determine if any approach to quantum computing can achieve utility-scale operation by 2033. QBI will use a multi-stage approach, Stages A, B, and C, to assess the proposed concepts, with each stage representing increased scrutiny. Rigetti will advance to Stage A, a 6-month performance period focused on the Company’s utility-scale quantum computer concept worth up to $1 million upon completing program milestones.
Developed by Boulder Imaging, IdentiFlight is a stateof-the-art bird detection and informed curtailment system designed to mitigate the impact of wind energy projects on bird populations. Using advanced artificial intelligence and optical sensor technology, IdentiFlight detects, classifies, and quantifies risks to sensitive bird species such as the Cape vulture, martial eagle, Verreaux’s eagle, and blue crane. If collision risk meets a predetermined threshold, IdentiFlight provides temporary shutdown on demand to impacted wind turbines – optimizing energy production while minimizing environmental impact.
The first South African IdentiFlight station was installed in June 2024. This system has already been trained to detect and classify more than 20 unique species with over 98 percent accuracy, including the Cape vulture, African marsh har rier, and African fish eagle.
Rigetti’s proposed concept for designing and building a Utility-Scale Quantum Computer (USQC) combines the Company’s proprietary multi-chip architecture with scalable quantum error correction (QEC) codes. Rigetti’s long-time partner and leader in QEC technology, Riverlane, will collaborate on this project and bring their expertise to help refine the proposed USQC concept and validate the underlying technology. DARPA defines a USQC as a system whose computational value exceeds its costs. The dominant factors in the scalability and speed of a USQC are the physical qubit architecture and QEC code. A common QEC technique is surface code, which
“Rigetti has spent the last decade developing the IP and expertise needed to build and deliver high-performing quantum computers. The DARPA QBI sets out to prove a realistic path to quantum utility, which we believe we are well positioned to deliver,” says Dr. Subodh Kulkarni, Rigetti CEO. “The DARPA QBI program is closely aligned with Rigetti’s technology roadmap, including building our QEC capabilities and developing fault-tolerant architectures.”
Rigetti’s quantum computers are based on superconducting qubits, which are a leading modality due to their fast gate speeds and the ability to leverage well-established semiconductor manufacturing techniques. The Company has made several breakthroughs in the superconducting qubit industry that will be the backbone of its roadmap to higher qubit count systems. This includes a novel
“Both the performance of our currently deployed 84-qubit Ankaa-3 system at 99.0% ISWAP gate fidelity and our bold roadmap taking us to 100+ qubits with a 2x reduction in error rates along with qLDPC code puts us in a strong position to meet DARPA’s 2033 target,” says David Rivas, Rigetti CTO.
Rigetti’s strategy to achieve a USQC focuses on implementing an efficiently-scaling qLDPC code in hardware. This will deliver the best of both worlds: the fastest quantum computing modality running the most efficient codes natively. Rigetti plans to combine these technologies and manufacture quantum integrated circuits capable of natively running qLDPC codes and reaching the high qubit counts and performance metrics needed for utility-scale applications.
Inside Track
Allen-Vanguard delivers its latest EW product – RF Decoy (TURMOIL) to a NATO nation.
Allen-Vanguard, the global leader in providing customized solutions and enabling technology across the Cyber and Electromagnetic Activities (CEMA) domain, has just announced the ahead-of-scheduled delivery of its latest RF Decoy product, TURMOIL, to a long-standing NATO nation. Due to the nature of the technology and its potential roles, Allen-Vanguard cannot disclose more specific details. Traditionally known for its RF Detect and Defeat capabilities, especially against Unmanned Air Systems (UAS) and Radio Controlled Improvised Explosive Device (RCIED) threats employed by terrorists and extremists, Allen-Vanguard has deep expertise across the whole CEMA domain and has rapidly developed this new disruptive technology to
meet a specific market demand and customer requirement. This is a shift for Allen-Vangauard into EW activities, and the team is excited by the potential being identified in the modern Electro Magnetic Spectrum Operational (EMSO) landscape.
Allen-Vanguard’s TURMOIL RF Decoy is an electromagnetic countermeasure that delivers a tactical advantage against adversaries by leveraging electromagnetic warfare (EW) capabilities. TURMOIL creates false targets by accurately emulating friendly forces’ RF signatures and increasing electromagnetic activity to distract, disrupt, and deceive during an enemy’s targeting phase. TURMOIL aids freedom of maneuver in the electromagnetic space and creates time and space for formations to deliver surprise offensive operations.
Bobby Strawbridge, President of Allen-Vanguard, said:
“It is an exciting period for Allen-Vanguard as we harness the vast technical experience and expertise we possess in the Company to meet the CEMA demands of the modern battlefield. I am extremely pleased that Allen-Vanguard has entered the EW arena with this new product, not only because it meets a customer’s specific requirements but also because it demonstrates how we at Allen-Vanguard can quickly and effectively apply our existing knowledge and skills to deliver success in new areas of operation.”
Pixus and COMINT Consulting Collaborate to Enhance Ruggedized Emerson/NI SDRs. New Signal Decoding Software Option Boosts Capabilities
Pixus Technologies, a provider of embedded computing and enclosure solutions, offers ruggedized versions of Emerson (Ettus Research brand) USRP Software-Defined Radios (SDRs). COMINT Consulting provides superior signal decoding and analysis software/firmware for SIGINT and COMINT. The combined solution provides tailored waveform monitoring/control options for a wide range of applications.
The Pixus-ruggedized Emerson SDRs in the RX310 and RX410 versions are ideal for Electronic Warfare, SIGINT, COMINT, and other applications. Working seamlessly with the Krypto1000 VHFSHF and satellite software suite, the SDRs can perform all of the functions of the SIGINT mission: Control, Collection, Precision Classification, Demodulation and Decoding, Parsing and even Decryption with the KeyFinder option.
The SDRs are designed to allow the units to be used in wider temperature ranges up to –40C to +71C, are fully IP67 sealed, and are areas with higher shock and vibration. The ruggedized SDRs have been qualified to meet MIL810 and MIL461. Pixus also offers the NI B210, N310, X310, and other SDRs in IP67 weatherproof and MIL rugged formations. The Company also provides small form factor (SFF) customized designs per customer specifications and VITA standard architectures.
Inside Track
Northrop Grumman to Equip Australian C-130J Fleet with Radio Frequency Countermeasures System
The Royal Australian Air Force (RAAF) has selected Northrop Grumman Corporation to provide the AN/ ALQ-251 advanced radio frequency countermeasures system for its C-130J airlifter fleet.
The AN/ALQ-251 protects against advanced weapon systems and agile electronic warfare systems in dense electromagnetic spectrum environments.
The system gives aircrews full situational awareness, radar warning, and precision direction in finding radio frequency threats.
RAAF C-130J integration takes advantage of commonality with a current United States Department of Defense program of record, reducing lifecycle costs and enabling interoperability.
Northrop Grumman to Equip Australian C-130J Fleet with Radio Frequency Countermeasures System.
The Royal Australian Air Force has selected the AN/ALQ-251 radio frequency countermeasures system for its C-130J fleet.
Expert: James Conroy, vice president of navigation, targeting, and survivability at Northrop Grumman: “With the increase of advanced radio frequency threats, our digital AN/ALQ-251 system allows the RAAF’s C-130J fleet the freedom of maneuver to carry out its wide range of humanitarian and national security missions safely.”
Details on AN/ALQ-251:
The Northrop Grumman AN/ALQ-251 is a radio frequency countermeasures system that provides radar warning, threat identification, geolocation, and countermeasure capabilities, with the flexibility to adapt to future threats. It is currently in production for the US Department of Defense and is available to international partners.
The AN/ALQ-251 is part of Northrop Grumman’s family of electronic warfare systems, which scale to virtually any platform or mission, including current and future platforms. Northrop Grumman’s sensors and electronic warfare systems give warfighters superiority across the spectrum and allow for faster, more informed decision-making.
Inside Track
Anduril and Microsoft are partners in advancing the US Army’s Integrated Visual Augmentation System (IVAS) program.
Microsoft Corp. and Anduril Industries, a leader in defense technology, announced an expanded partnership to drive the next phase of the US Army’s Integrated Visual Augmentation System (IVAS) program. Through this partnership agreement and pending Department of Defense approval, Anduril will assume oversight of production, future hardware and software development, and delivery timelines. This agreement also establishes Microsoft Azure as Anduril’s preferred hyperscale cloud for all workloads related to IVAS and Anduril AI technologies.
As warfare grows more complex, the need for intuitive, information-rich tools becomes increasingly urgent. Traditional mission command systems — that rely on flat maps, static dashboards, and verbal communication — are ill-suited to the dynamic nature of modern battlefields. The IVAS program represents a groundbreaking step forward in military technology, providing soldiers with a comprehensive, body-worn system that integrates advanced augmented reality (AR) and virtual reality (VR) to give warfighters beyond line-of-sight perception capabilities, increasing combat effectiveness, survivability against drones, and accelerating mission command of unmanned systems.
This collaboration between Anduril and Microsoft leverages the distinct strengths of each Company, ensuring continuity in the program while advancing IVAS goals of enhancing combat effectiveness, awareness, and decision-making on the battlefield. Anduril’s mission focuses on innovation in defense technology, a deep understanding of military requirements, and a unique approach to manufacturing defense products. These will ensure future program development specifically tailored to the evolving needs of the Army, as well as production at scale and lower unit cost.
Microsoft’s advanced cloud infrastructure and AI capabilities will continue to provide a robust backbone for the program, enabling seamless data integration and real-time insights critical to soldier effectiveness. Artificial intelligence will be foundational to all technical innovations for national security missions, and the cloud is essential to successfully delivering AI whenever and wherever it is needed. Anduril will establish Azure as its preferred hyperscale cloud to support its AI development through this agreement. Through its commercial, US government, and classified clouds, Azure provides high resiliency, sophisticated capabilities, flexibility, and advanced security designed to meet the stringent compliance requirements of the nation’s most sensitive data.
Since its founding in 2017, Anduril has focused exclusively on developing advanced defense technologies to solve unique challenges for warfighters. Palmer Luckey, Anduril’s founder, who previously revolutionized the VR industry with Oculus, brings unparalleled expertise and vision to this effort.
“The IVAS program represents the future of mission command, combining technology and human capability to give soldiers the edge they need on the battlefield,” said Palmer Luckey, founder of Anduril Industries. “The ultimate goal is to create a military ecosystem where technology acts as an extension of human capability. By empowering soldiers with the tools they need to make faster, smarter decisions, we’re building a future where technology and human ingenuity combine to ensure mission success.”
“We are incredibly proud of the work our teams have put into helping the US Army transform its concept of a soldier-borne AR headset into reality with the IVAS program. Our Soldier-Centered Design approach helped reimagine technology development with the Army that continuously took in real-world soldier feedback to develop a product that soldiers love,” said Robin Seiler, corporate vice president of mixed Reality at Microsoft. “We are excited to partner with Anduril for the next phase of IVAS and leverage our combined strengths to meet our commitments on this vital program and deliver a game-changing capability for every U.S. soldier.”
Microsoft and Anduril have previously collaborated to integrate Anduril’s Lattice platform into IVAS to provide rapid AI-enabled situational awareness capabilities to enhance soldier safety and operational effectiveness. This new agreement brings Microsoft and Anduril closer together to maximize their unique expertise in cloud computing, artificial intelligence, and defense manufacturing to enhance technical developments and operational efficiencies further.
Inside Track
ManTech Paves the Way in Secure AI Adoption with Google Cloud for Secure Data
ManTech announced that it is operationalizing Google Cloud’s Vertex AI platform and Gemini models within a Controlled Unclassified Information (CUI) environment at Cybersecurity Maturity Model Certification (CMMC) Level 2. This achievement marks a significant step forward in the responsible and secure adoption of advanced AI for government and defense applications.
“We are dedicated to providing innovative AI solutions that equip our employees and clients with the tools to excel in a cyber world rapidly evolving at unprecedented technological speeds,” said Mike Uster, ManTech’s Chief Technology & Information Officer.
“ManTech’s deployment of Google Cloud’s Vertex AI and Gemini in a secure CMMC Level 2 environment is a major step forward in bringing the power of responsible AI to the public sector,” said Tony Orlando, General Manager Specialty Sales, Google Public Sector.
“By leveraging advanced capabilities for large-scale data analysis and streamlined operations, this collaboration unlocks the potential for groundbreaking solutions that drive innovation and efficiency for government agencies.”
ManTech’s implementation of the Vertex AI platform and Gemini models focuses on leveraging its advanced capabilities for tasks such as: Large-scale data analysis: Gemini models on Vertex AI’s one million token context window for the processing and analysis of vast quantities of data, enabling deeper insights and more informed decision-making. Streamlined operations: Automating tasks such as report generation and data summarization frees valuable time and resources for innovation versus routine tasks.
Security at the Forefront of AI Adoption
ManTech has prioritized the secure implementation of the Vertex AI platform and Gemini models, addressing
key requirements for AI deployment in sensitive environments, including:
• Data Encryption: All CUI data processed by the Gemini models are encrypted in transit and at rest, ensuring its confidentiality and integrity.
• Source Data Citation: Gemini models clearly cite the source data used in their responses, enabling traceability and validation of information.
• Proprietary Information Protection: Strict access controls and data governance policies are in place to safeguard ManTech’s and its clients’ proprietary information.
“This achievement by ManTech accelerates the adoption of advanced AI technologies within secure government and defense environments,” said ManTech Chief Innovation Officer Eric Brown. “ManTech sets a new standard for responsible AI implementation in the public sector by prioritizing security and compliance.”
Cross Domain Solutions in the Age of AIDriven Cyber Attacks:
By John Reardon, Editor-in-Chief, COTS Journal
In the quiet corridors of modern cyber warfare, a new kind of arms race is playing out. Artificial Intelligence has opened the door to a new frontier in both attack and defense. This is not science fiction. This is now.
The threats facing high-assurance cybersecurity systems are evolving faster than ever—changing shape, disguising intent, and learning from failure. The modern battlefield is no longer just kinetic; it is digital, intelligent, and always on. This is the context in which two powerhouses— Everfox, with 25 years of experience in cross-domain cybersecurity, and Palantir Technologies, the data analytics giant trusted by the Department of Defense—have joined forces. Their mission? To take the most advanced AI tools and the most resilient cybersecurity frameworks, and fuse them into a single, real-time defense mechanism for the nation’s most sensitive classified networks.
This partnership isn’t just a meeting of minds; it’s a practical answer to the growing complexity of securing top-secret data in an age when attackers are using AI to probe, learn, and evolve in real time. And if that’s how the attackers are thinking, then the defenders must think one step ahead.
When Intelligence Fights Back
An Intelligence War with No Front Lines
Picture this: an AI-driven adversary is given a simple goal—breach a network. It’s deployed onto the internet with a high-performance compute platform behind it and an evolving attack surface built on machine learning. It doesn’t need sleep. It doesn’t make the same mistake twice. It keeps going, adjusting, rewriting its code, identifying patterns, until—eventually—it gets in.
That’s the level of sophistication we’re facing. According to George Kamis, CTO at Everfox, this reality is the reason why cybersecurity can no longer rely on static defenses. “Our solutions need to evolve just like the threats. Think of it like an airport checkpoint,” Kamis explains. “You have a firewall—like the initial ID check—and then layers of deeper inspection, like TSA screenings. Except in our case, it’s not just people, it’s petabytes of data flowing across domains.”
Everfox’s cross-domain solutions are built with this layered security concept in mind. Much like how a TSA agent identifies suspicious behavior, the system analyzes and filters data in motion—watching for anomalies, identifying threats, and stopping them before they can act. It’s real-time, it’s proactive, and it’s designed not just to detect threats but to thwart them before they take hold.
The Mission Manager Edge
Where Everfox brings hardened cross-domain solutions to the fight, Palantir brings the AI muscle. Their Mission
Manager platform leverages powerful machine learning to sift through vast troves of real-time and historical data, flagging threats, connecting the dots, and learning patterns that would take human analysts weeks—or months—to understand.
By combining Everfox’s domain-specific filtering with Palantir’s pattern-recognition prowess, mission-critical classified networks can now incorporate open-source intelligence securely. That’s a massive leap forward. In the past, security concerns made it virtually impossible to mingle open and classified sources without a laborious, manual vetting process. Now, the system itself can decide what’s safe, what’s actionable, and what needs immediate attention.
Real-World Defense, Not Dashboard Warnings
“This is not about generating more red alerts on a dashboard,” Kamis says. “This is about acting. Automatically. Intelligently. And fast.”
One of the most powerful techniques now used in this joint effort is something known as CDR—Construction, Deconstruction, and Reconstruction. It works like this: data streams are broken down to their most basic elements. Think of stripping a car down to the bolts, inspecting every part for tampering, then rebuilding it from scratch. That’s how CDR ensures no malware, backdoors, or embedded threats make it into a secure domain. It’s surgical, precise, and utterly necessary in today’s threat landscape. And that’s only the beginning.
Seamless Cross-Domain Data Without Air Gaps
In military parlance, the goal today is Joint All-Domain Command and Control—a unified approach to warfare where information flows seamlessly between air, land, sea, space, and cyber domains. But real-time information sharing across classification levels has traditionally been stifled by air-gapping and sandboxing techniques, which are secure but slow.
That’s where Everfox’s cross-domain capabilities shine. Their tools can strip sensitive information from raw data and deliver sanitized, actionable intelligence faster than ever before. This means mission-critical decisions—such as identifying a drone threat, tracking an adversary’s position, or deploying a cyber countermeasure—can happen in seconds, not hours.
The result? Operators on the edge of the battlefield now have confidence that the data they’re acting on is not only timely, but clean, trusted, and secure.
From Commercial Battlefield to Military Lines
What gives Everfox such an edge in this space is its roots in commercial cybersecurity. Long before this partnership, Everfox was battling advanced persistent threats (APTs) in sectors like finance, energy, and healthcare—environments where the stakes are high, and the adversaries are relentless. These hard-won lessons in the private sector now feed directly into the military domain. Every attempted breach, every novel exploit, every AI-powered probe adds to a growing library of knowledge. And with Palantir’s AI integrating that
knowledge into adaptive learning models, the joint solution doesn’t just resist attacks—it evolves with them.
Looking Ahead: The Future of AI in Defense
The Department of Defense has been clear: the future of national security depends on both cross-domain data access and full-spectrum AI implementation. These goals are inseparable.
The historical parallels are striking. During World War II, Alan Turing’s efforts at Bletchley Park to crack the Enigma code didn’t just help win battles—they saved lives. In much the same way, today’s digital warfare demands new machines, new codebreakers, and new partnerships to keep pace with adversaries who are every bit as cunning and technologically enabled.
Cybersecurity is no longer just about walls and gates. It’s about sensors, instincts, and the ability to think like your attacker—only faster. Everfox and Palantir’s collaboration embodies this shift. It’s not just another security product or AI algorithm. It’s a blueprint for how we fight and win in the digital domain.
In a world where threats are measured in milliseconds and battle lines exist in code, the question isn’t whether AI will be used in cyberwarfare. It already is. The real question is: who’s using it better?
With the combined power of Everfox and Palantir, the answer—for now—is clear.
Precision Timing at the Heart of Defense Open Architecture Systems
By Odile Ronat, SiTime, Director of Product Marketing
Timing plays a central role in ensuring that electronic systems operate reliably and harmoniously. This is never more critical than in the battlefield. Electronic systems such as unmanned vehicles, ground sensors, command centers and tactical communications must work in coordination. When multiple platforms are deployed in joint operations—for functions such as real-time drone video, jamming enemy signals or data communication —they must be synchronized. This ensures data transmission is cohesive for reliable and timely decision making and action.
If various subsystems are provided by different manufacturers and introduced at different times are not synchronized, it complicates defense operations. As the industry moves toward open architecture implementations, timing misalignments can lead to latency issues, degraded performance or, in the worst cases, system failure. This article discusses how precision timing helps ensure interoperability with precisely synchronized data transmission.
Reduces logistics tails by enabling common sparing. Eliminates the need for end-of-life buys for 30+ years sustainment by enabling hardware modernization every 5-10 years. (Source: UA Army DEVCOM C5ISR Center)
For systems built on modular open standards, the system requirements for communications, command and control, electronic warfare, intelligence gathering and others, dictate the timing device requirements. Key considerations include:
• Size, weight, power and cost (SWaP-C) reduction: Military platforms, particularly mobile and airborne systems, require compact, lightweight and energy-efficient components. Design engineers also prefer readily available, off-the-shelf components with reasonable lead time and continuity of supply.
• Shock, vibration and temperature resilience:
Reliable and stable frequency is essential to maintain optimal performance. Even minor variations can severely degrade system accuracy and functionality. Military systems must withstand harsh environments. These include changing and extreme temperatures, high vibration levels and severe shock conditions. Timing components must be robust and reliable
• Holdover performance:
GPS is the source of reference timing in over 700 DoD weapon systems. Unfortunately, GPS is vulnerable to attacks. Maintaining accurate timing even when a GPS signal is unavailable or untrustworthy is critical for successful military operations.
Precision timing devices are distributed throughout electronic systems and must perform in a wide variety of contexts to ensure that all subsystems remain synchronized. The ability to maintain functional and accurate time in complex and high-stakes defense environments can make the difference between mission success and failure.
Why MEMS Timing Is the Right Choice for Defense Applications
Micro-electro-mechanical systems (MEMS) silicon timing technology is a means for solving some critical defense challenges. Ruggedized Precision MEMSbase oscillators such as temperaturecompensated oscillators (TCXOs) and ovencontrolled oscillators (OCXOs) provide significant advantages over quartz-based products. These improvements include:
• SWaP reduction:
COTS Ruggedized MEMS oscillators are proven low SwaP solutions. For instance, the SiT5543 in a 7 mm x 5 mm package consumes only 135 mW, making it an ideal choice for platforms where SwaP constraints are a significant concern. This device can replace a quartz OCXO that would consume almost 4X the energy at 500 mW in a footprint 10X to 20X larger.
• Frequency stability over temperature:
MEMS timing technology is advancing, continuing to provide a range of options to replace quartz TCXOs and OCXOs. SiT5348 MEMS Super-TCXO is rated at +50 ppb over a wider temperature range of up to -40 to 105°C with repeatable and predictable frequency with no frequency dip or activity jump. New SiT5543 TCXO rated at +5 ppb over -40 to 95°C is the only TCXO that achieves this temperature stability with lower power, smaller size and higher reliability than a quartz OCXO. The new SiT7111 oven-controlled MEMS oscillator delivers + 1 ppb over -40 to 95°C, a wider temperature range than most quartz OCXOs.
• Shock and vibration resilience: Ruggedized MEMS oscillators have far lower sensitivity to vibration, typically around 1E-11/g maximum, a level not achieved in commercially available quartz oscillators. These MEMS oscillators can survive shocks of 30,000g and beyond, with simple construction and reliably: this level of shock survivability is a big challenge even for customized, purpose-built quartz oscillators. This makes MEMS ruggedized oscillators well-suited for military systems deployed in extreme conditions. Unlike quartz oscillators, MEMS oscillators are not reliant on the size of the crystal, which allows them to be more compact and more resistant to vibration and shock. MEMS oscillators offer greater reliability, with a meantime between failure (MTBF) of over 2 billion hours. This is a significant improvement over quartz oscillators.
• Holdover performance:
MEMS oscillators, such as the SiT7111, deliver superior holdover performance with a repeatable typical time error of ±1 microsecond over a 24-hour period, ensuring that systems remain accurate even in GPS-denied environments, and a superior performance compared to quartz OCXO.
• Easier design integration:
Low SwaP MEMS oscillators with low sensitivity to board level temperature swings, vibrations and power supply noise, and without the need for external voltage regulators, simplify layout and thermal management. This makes them easier to design into systems that require robust precision timing. MEMS oscillators are designed with I2C or SPI frequency tuning options, allowing for greater flexibility in their application.
MEMS Precision Timing Improves Open Architecture System Performance
The characteristics of MEMS-based precision timing devices make it ideal for a wide range of defense applications, including:
• GPS disciplined oscillators (GPSDO): Ruggedized MEMS TCXOs with superior repeatable frequency stability without frequency jumps enables fast, robust and reliable signal decoding, including encrypted signals with long integration time, continuous signal lock under rough conditions, fast signal reacquisition time when signal is lost and improved resistance to jamming and other interferences.
• Assured positioning, navigation, and timing (A-PNT):
In environments where GPS signals are denied or degraded, the local oscillator becomes the timing reference and needs to maintain accuracy from minutes to hours to days—a condition referred to as holdover. The device of choice will have to have a limited drift from the time the GPS synchronization is lost, over time and under changing environmental conditions. New ultra-stable SiT7111 MEMS oscillator can replace expensive and fragile chip-scale atomic clocks (CSACs) in A-PNT modules in a smaller form factor with extended reliability.
• Software-defined radio: For software-defined radios, which require high levels of phase noise and frequency stability, new MEMS Super-TCXOs, such as SiT7201 and SiT7202, offer the necessary
performance in a SWaP-C package and simplify quartz TCXO-based circuitry or eliminate the need for an OCXO. In addition, their phase noise is virtually insensitive to vibrations, ensure continuity of performance and signal even through the roughest environments.
• Radial clock and switch cards: Radial clock cards and switch cards are critical in distributing timing signals across various modules to ensure that all parts of a system operate in sync, whether through Precision Time Protocol (PTP) such as IEEE1588 or Synchronous Ethernet (SyncE). The performance of these cards directly affects the reliability and effectiveness of the overall system. The latest MEMS oscillators, with their superior environmental resistance and overall stability, are well-suited to meet the demands in synchronization of these critical system functions.
COT’S PICKS
Russelectric, A Siemens Business, Highlights Advanced Microgrid Controls Solution for Enhanced Power Resiliency in Critical Sectors
Offers significant benefits, including energy cost reduction, carbon footprint minimization, and improved business continuity Russelectric, A Siemens Business, highlights its Advanced Microgrid Controls
Solution, an integrated hardware and software platform designed to optimize facility power management and enhance resiliency in critical sectors. This comprehensive solution may present significant benefits, including energy cost reduction, carbon footprint minimization, and improved business continuity during grid outages.
The Advanced Microgrid Controls Solution is a modular, scalable, and seamless integration of hardware and software components. At its core, the system utilizes Russelectric’s transfer switches, switchgear, and power controllers to integrate a facility’s onsite demand and generation assets comprehensively. The solution connects, controls, and optimizes facility generation assets such as generators, battery backup, photovoltaic arrays, and loads, benefiting from a wide range of distributed energy resource (DER) technologies.
Teledyne FLIR OEM Launches Prism
SKR Automated Targeting and Homing Software
Prism SKR addresses the ever-growing need for autonomously guided weapon systems: low-cost missiles, loitering munitions, and one-way drones.
Teledyne FLIR OEM, part of Teledyne Technologies Incorporated, announced Prism™ SKR, the newest Prism embedded software ecosystem addition. Prism SKR, pronounced “seeker,” sets a new standard in precision automatic tar-
Ideal for data centers, Russelectric’s Advanced Microgrid Controls Solution improves business continuity during grid outages through dynamic islanding capabilities, fast and reliable decoupling, automatic black
get recognition (ATR) for autonomously guided weapon systems, including loitering munitions, air-launched effects (ALE), counter-unmanned aerial systems (C-UAS), low-cost missiles, and smart munitions.
Prism SKR operates seamlessly at the edge of low-power embedded processors. Compatible with infrared (IR) thermal and visible camera data, Prism SKR’s advanced algorithms provide real-time target position, identification, direction of motion, and aim-point localization. The target information guides the platform’s
March 2025
COT’S PICKS
embedded software ecosystem and Teledyne FLIR IR camera modules.”
The Prism software ecosystem streamlines development and shortens time to market with out-of-thebox compatibility. Developers and integrators can utilize detection and tracking models leveraging Teledyne FLIR OEM’s six million visible and thermal annotations, starting with the Prism AI embedded software. Custom target lists can quickly be incorporated using the Prism AIMMGen synthetic dataset and model training
toolchain. Finally, Prism Supervisor empowers autonomous mission control and platform guidance with Prism SKR, where ATR is required.
With its advanced features, robust ecosystem, and a complete application programming interface (API) and software development kit (SDK) for customization and performance optimization, Prism SKR is the ultimate solution for OEMs seeking to enhance the precision and reliability of autonomous weapon systems.
start of backup generators, and seamless grid resynchronization. Additionally, the solution helps reduce energy costs and minimize carbon footprint while improving facility power quality.
autonomous flight system for air-toground (A2G), ground-to-air, air-to-air, and ground-to-ground operations.
“Compatible with existing autonomy and navigation software via a Weapon Open System Architecture (WOSA) compliant interface, Prism SKR represents a significant leap forward in autonomous precision target detection, identification, and tracking technology for defense OEM developers and integrators,” said Dan Walker, vice president of product management, Teledyne FLIR OEM “Additionally, developers can reduce development risk and costs with plug-and-play integration with the Prism
VarTech Systems Launches the PowerCube (3rd generation)
A High-Performance Industrial Small Form Factor Computer Designed for Space-Constrained Areas
VarTech Systems Inc. has announced the release of the PowerCube (3rd Gen.)—a compact and powerful small
form factor computer (SFFC) built for rugged and industrial applications. Designed for space-constrained environments in industrial and mobile applications, the new Gen3 PowerCube delivers enhanced processing power, improved thermal efficiency, and a smaller, ultra-compact form factor.
The PowerCube is engineered to with-
stand extreme temperatures, shock, and vibration, ensuring reliable operation in numerous industrial settings, including manufacturing, industrial automation, energy, transportation, and military applications.
With customizable I/O configurations, high-performance CPUs, and solid-state storage, it seamlessly integrates into demanding applications where space is limited, but performance is non-negotiable. PowerCube options include integrated power supplies, removable storage drives, and shock mounting provisions.
“The third generation of the PowerCube sets a new standard for customizable, industrial small form factor computers,” said Richard Lachney, President at VarTech Systems. “Its slimmer, more efficient design provides cutting-edge performance while maintaining the rugged durability our customers expect.”
COT’S PICKS
Cirrascale Cloud Services Debuts Next-Generation Inference Platform Aimed at Deploying Customer Models on Ideal Accelerators
The new inferencing platform can comprehensively analyze AI models, deploy them on the most ideal AI accelerators, and dynamically balance those workloads across multiple regions.
March 2025
COT’S PICKS
OSS Unveils the Ultra-Dense 16-Way GPU Expansion System. The 6U, 16-way Ponto breaks new ground in the GPU expansion market.
Cirrascale Cloud Services announced the early preview of its groundbreaking Cirrascale Inference Platform—an enterprise inference-as-a-service solution engineered to optimize AI model performance and cost for high token volume applications.
Specifically designed for large-scale model deployment, the platform will launch with the NVIDIA Blackwell architecture, featuring the NVIDIA HGX B200 and NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs. The platform automatically selects the optimal AI accelerator to adapt to changing model requirements,
user demands, and workflow shifts, ensuring unmatched performance and substantial cost savings compared to traditional methods. Additionally, it dynamically balances workloads across regions, helping to smooth out peak demands, enhance operational efficiencies, and reduce costs
“Enterprise AI adoption is accelerating, and while hyperscalers are suitable for initial validation and model training, they can introduce additional complexities and costs with larger volume usage,” said Nick Pandher, vice president of product at Cirrascale Cloud Services. “With the launch of the Cirrascale Inference Platform, customers unlock access to a high-performance, seamlessly deployable cloud inference solution that intelligently optimizes their AI configurations for both efficiency and cost-effectiveness.”
Engineered with enterprise-grade secu-
rity, the Cirrascale Inference Platform features isolated pipelines that integrate seamlessly with both hyperscalers and on-premises environments. Leveraging dedicated bare-metal performance eliminates the risks of multi-tenancy common to AI model pipelines delivered through virtualized infrastructure, ensuring high token throughput and robust security. The platform is designed to simplify management and scaling, allowing enterprises to maintain low-volume models on-premises while offloading demanding workloads to a high-performance, cost-efficient solution—complete with advanced features like Retrieval Augmented Generation (RAG) for enhanced model performance.
“As we moved our AI models to scaled inference, we were looking for a cloud services provider that optimized for the latest AI accelerators that offer low-precision workloads while providing predictable pricing,” said Martin Woodall, CEO at DroneData. “Our robotics solutions require fast and predictable inference with real-time results. The new FP4 technology for inference was critical to get the performance we need.”
“The NVIDIA Blackwell architecture delivers exceptional performance and scalability for AI inferencing,” said Dave Salvator, director of accelerated computing products at NVIDIA. “The new Cirrascale instances with NVIDIA HGX B200 as the leading system for this platform will empower enterprises to seamlessly deploy and optimize large-scale AI models with efficiency and cost-effectiveness.”
The platform will be available across multiple U.S. regions and select international regions to reduce latency and optimize performance for hyperscaler and on-premises application workflows. Early access launches in April, with wider access available in the summer.
One Stop Systems, Inc. unveils the world’s first PCIe Gen5 expansion system, Ponto, which supports 16 675 W full-size add-in cards in a 6U rack space, paving the way for the next generation of GPU and accelerator devices.
Ponto, an ultra-high-power density PCIe expansion system, supplies 11 kW of power to expansion slots, making it the first product in its class. This is equivalent to an unparalleled 1.83 kW per rack unit, which enables users to pool 16 dual-width add-in cards of up to 675 W, such as the 600 W H200 NVL, in one system. The Ponto is also configurable to support 32 single-width add-in cards.
With Ponto, OSS leveraged its expertise in thermal management to solve the challenging problem of cooling the industry’s most potent expansion system. High-voltage fans, specialized ducting, and dynamic fan controls ensure that the system’s GPUs will run at full power, full time.
OSS’s Unified Baseboard Management Controller (U-BMC) enables advanced system monitoring that integrates seamlessly with the host server’s system management interface. U-BMC manages, monitors, and controls the telemetry of OSS products with advanced capabilities. This software securely enables OSS customers to unify their
systems management and enjoy unique benefits such as dynamic fan controls, GPU power throttling, and add-in card telemetry.
Ponto will prove invaluable in data center environments requiring high-density compute performance for high data throughput applications and is well-suited for PCIe expansion-focused composable infrastructure architectures.
Supermicro Adds Portfolio for Next Wave of AI with NVIDIA Blackwell Ultra Solutions, Featuring NVIDIA HGX™ B300 NVL16 and GB300 NVL72
Air- and Liquid-Cooled Optimized Solutions with Enhanced AI FLOPs and HBM3e Capacity, with up to 800 Gb/s Direct-to-GPU Networking Performance.
Supermicro, Inc., a Total IT Solution Provider for AI, Cloud, Storage, and 5G/Edge, is announcing new systems and rack solutions powered by NVIDIA’s Blackwell Ultra platform, featuring the NVIDIA HGX B300 NVL16 and NVIDIA GB300 NVL72 platforms. Supermicro and NVIDIA’s new
AI solutions strengthen leadership in AI by delivering breakthrough performance for the most compute-intensive AI workloads, including AI reasoning, agentic AI, and video inference applications.
“At Supermicro, we are excited to continue our long-standing partnership with NVIDIA to bring the latest AI technology to market with the NVIDIA Blackwell Ultra Platforms,” said Charles Liang, president and CEO of Supermicro. “Our Data Center Building Block Solutions® approach has streamlined the development of new air and liq-
uid-cooled systems, optimized to the thermals and internal topology of the NVIDIA HGX B300 NVL16 and GB300 NVL72. Our advanced liquid-cooling solution delivers exceptional thermal efficiency, operating with 40% warm water in our 8-node rack configuration or 35% warm water in a double-density 16-node rack configuration, leveraging our latest CDUs. This innovative solution reduces power consumption by up to 40% while conserving water resources, providing both environmental and operational cost benefits for enterprise data centers.”
COT’S PICKS
NVIDIA’s Blackwell Ultra platform is built to conquer the most demanding cluster-scale AI applications by overcoming performance bottlenecks caused by limited GPU memory capacity and network bandwidth. NVIDIA Blackwell Ultra delivers an unprecedented 288GB HBM3e of memory per GPU, drastically improving AI FLOPS for AI training and inference for the largest AI models. The networking platform integration with NVIDIA Quantum-X800 InfiniBand and Spectrum-X™ Ethernet doubles the compute fabric bandwidth, up to 800 Gb/s.
Supermicro integrates NVIDIA Blackwell Ultra with two types of solutions: Supermicro NVIDIA HGX B300 NVL16 systems, designed for every data center, and the NVIDIA GB300 NVL72, equipped with NVIDIA’s next-generation Grace Blackwell architecture.
Supermicro NVIDIA HGX B300 NVL16 system
Supermicro NVIDIA HGX systems are the industry-standard building blocks for AI training clusters. They have an 8-GPU NVIDIA NVLink™ domain and a 1:1 GPU-to-NIC ratio for high-performance clusters. Supermicro’s new NVIDIA HGX B300 NVL16 system builds upon this proven architecture with thermal design advancements in both a liquid-cooled and air-cooled version. For B300 NVL16, Supermicro introduces a brand new 8U platform to maximize the output of the NVIDIA HGX B300 NVL16
Curtiss-Wright Announces Expanded Capabilities of IQ-Core Software Suite for Tactical Network Management
The latest updates to IQ-Core Software add numerous new capabilities, improve ease of use, and improve network security at the tactical edge.
Curtiss-Wright’s Defense Solutions Division has announced the latest enhancements to its industry-leading IQ-Core® Software suite of network management solutions.
board. Each GPU is connected in a 1.8TB/s 16-GPU NVLink domain, providing a massive 2.3TB of HBM3e per system. Supermicro NVIDIA HGX B300 NVL16 improves upon performance in the network domain by integrating 8 NVIDIA ConnectX®-8 NICs directly into the baseboard to support 800 Gb/s node-to-node speeds via NVIDIA Quantum-X800 InfiniBand or Spectrum-X™ Ethernet.
IQ-Core Software, a PacStar® family product line, is a comprehensive management software application designed for secure, tactical, and distributed network management – ideal for edge networks in disconnected, intermittent, and limited (DIL) environments. IQ-Core Software enhances mobility in the field, offering rapid and simplified network setup and management tasks at the tactical edge, replacing hours of tedious, error-prone manual configuration with precise, easy-to-follow wizards that can be completed in minutes. IQ-Core Software is designed for use on a wide range of platforms, including manned/unmanned Ground Defense, Airborne, Naval, and Industrial platforms.
“We are committed to continually extending and
March 2025
COT’S PICKS
•A suite of powerful new features to enhance the management and maintenance of INTEGRITY™ Global Security (IGS) security appliances
Strong, multi-factor authentication mechanisms such as Common Access Card (CAC) and smartcards mitigate the rise of data breaches and infiltration attempts, which is critical to maintaining the integrity of sensitive communications and data. This strengthens security by ensuring that only authorized personnel can access sensitive communication networks. Secure, quick access to communication tools eliminates the need for cumbersome password management. Adding support for the industry-standard REST API enables IQ-Core Software customers to easily integrate with third-party platforms. It helps connect multiple disparate systems together in a system-of-systems approach and delivers on the US DOD’s desire for composable
software solutions. New API functions supported by IQ-Core Software include the ability to more fully integrate with HPE Aruba ClearPass Policy Manager, Secure tunnel (VPN) and alarm management for Aruba Orchestrator, and integration of Aruba products into Commercial Solutions for Classified (CSfC) environments.
High-density power modules for bridging 48V and legacy 12V power buses
building on the foundation of our powerful IQCore Software suite of network management solutions designed for cutting-edge tactical communications,” said Brian Perry, Senior Vice President and General Manager of Curtiss-Wright Defense Solutions Division. “IQ-Core Software is already the de facto standard for battlefield network management and configuration software, and our team keeps bringing the new features needed to keep pace with our customers’ missions.”
New IQ-Core Software Capabilities:
•Enhanced security via support for smartcard and CAC multi-factor authentication
•REST- based Application Programming Interfaces (API)
Astronics to Provide Advanced Frequency Converter for NASA-Boeing X-66 Demonstrator Aircraft
Astronics Corporation, a leading supplier of advanced technologies and products to the global aerospace, defense, and other mission-critical industries, announced today it has been selected to provide the Frequency Converter Unit (FCU) for the NASA and Boeing Transonic Truss-Braced Wing (TTBW) X-66 aircraft demonstrator. In addition to designing the FCU, Astronics will collaborate closely with project stakeholders to support ground and flight tests, slated to start in 2028.
The new suite of features to enhance the management and maintenance of INTEGRITY™ Global Security (IGS) security appliances significantly reduces operational complexity. It ensures operators can swiftly adapt to changing environments at the tactical edge. IQ-Core Software now delivers faster provisioning, improved flexibility in managing multi-site operations, and accelerated network recovery capabilities. This gives network operators the agility and resilience to maintain mission-critical connectivity, even in the most challenging environments DCM non-isolated
The FCU changes the output of the newly adapted 115VAC, 3-phase Variable Frequency Generator to 115VAC, 3-phase, constant frequency (400 Hz) for the aircraft’s main power buses. This capability enables the use of legacy onboard systems and loads that are qualified to 400Hz constant frequency power.
“Built on Astronics’ proven technology platform, the Frequency Converter Unit offers unparalleled efficiency and reliability, tailored to meet the unique demands of this groundbreaking aviation project. Being part of this collaboration with Boeing and NASA is an honor. We believe our advanced frequency conversion technology will play a critical role in the successful testing and operation of the X-66 demonstrator, helping to advance the goal of more sustainable aviation,” said Jon Neal, President of Astronics AES.
Vicor non-isolated, regulated DCDC converter modules operate from a 40 – 60V input, generate a regulated 12V output, and are adjustable from 10V to 12.5V. They support up to 2000W and can be paralleled to attain higher power levels. They are complete converter solutions requiring minimal external components, yielding even higher power density for the complete power delivery network. The DCM has established a new benchmark for power density with a low-height, surface mount Converter housed in Package (SM ChiP), resulting in a 6x reduction in size over a conventional converter. With Vicor’s proprietary high-frequency switching technology, these modules deliver high efficiency over all line and load conditions. They can be used to establish a 12V bus or with legacy downstream point-of-load (PoL) converters. They are ideally suited for bridging 48V and legacy 12V power buses in industrial, aerospace, defense, and high-performance computing applications.
Astronics is a long-term supplier of electrical power system products to Boeing in both the commercial and defense markets. The TTBW program marks a significant milestone in the aviation industry’s pursuit of sustainability, highlighting innovative technologies that support the industry’s goal of achieving net-zero emissions by 2050.
COT’S PICKS
Aitech Unveils World’s First AI-Enabled Picosatellite Constellation Platform Transforming Space-Based Intelligence for Military, Defense, Public Safety, Agriculture, and Infinite Other New Markets
Aitech, a leading provider of rugged embedded systems for military, aerospace, and space platforms, introduced the picosatellite (PicoSat) constellation platform, IQSat‚Ñ¢.
The IQSat is AI-enabled and includes Artificial Intelligence (AI) and Machine Learning (ML) from Intuidex, Inc. incorporated in Intuidex’s Watchman for Space‚Ñ¢ (W4S‚Ñ¢) product for detecting and providing insight on patterns and anomalies for a range of applications, including public safety, agriculture, military, defense, climate, biology and more.
Built on Aitech’s 30 years of electronics legacy and trillions of miles flown in space, the IQSat is small enough to be held in the palm of a hand and can be deployed in a constellation of five to thousands. Offering coverage previously unattainable, IQSat is powerful enough to deliver actionable patterns of life analysis rapidly and frequently from any desired point on Earth.
“Space-based infrastructure solves big problems and fuels big advancements on Earth. The new IQSat platform is integral to space accessibility by delivering low cost, rapid deployment constellations that provide access to actionable information quickly and frequently for infinite applications,” said Pratish Shah, U.S. general manager, Aitech. “Whether used for military and defense, environmental or agricultural applications, communications or scientific research, the flexibility, cost, and availability of a solu-
tion like IQSat have not existed before - providing more accessibility to the power of space.”
Leveraging Intuidex’s Watchman for Space product, the IQSat offers Higher-Order LowResource Learning‚ in support of rapid doit-yourself user-based modeling, detection and tracking, pattern of life and anomaly detection, and constellation operations capabilities that enable new uses across a range of markets:
Military and Defense: With a low probability of intercept and low probability of detection, IQSat can deliver critical information on threat locations, directions, and velocities in any area of interest at tactically significant update rates directly to the warfighter
Space Situational Awareness: With flexible sensor options and constellation configurations, IQSat brings low-cost capability to detect space debris to prevent space collisions and monitor orbital traffic, as well as detect changes in satellite trajectories or potential anti-satellite threats
Space Habitat: Placed in the vicinity of Space Habitats, IQSat can deliver temperature, radiation levels, and other data variables to evaluate structure integrity by detecting micro-meteorite impacts and material degradation or external surveillance to detect space debris collision threats
Public Safety: A low-cost LEO constellation of IQSats can provide rapid detection of remote areas to identify natural disasters such
as floods and fires; assist in remote search and rescue operations through detection of objects, locations, and movement directly to the rescuer; or monitor infrastructure surveillance of dams, bridges, and other structures
Agriculture: IQSat can assist farmers by delivering fast and frequent insight on soil, crops, disease, weather, and other critical variables to help improve crop output
Science and Research: A constellation of IQSats can provide researchers and scientists with a low-cost solution for tracking unique weather patterns, monitoring wildlife migrations around the world, tracking ocean currents and wildlife, and many more capabilities
The IQSat provides mission flexibility for quick and cost-effective constellation design, reducing delivery time and cost of custom sensing satellites and constellation configurations. Leveraging Watchman for Space, the IQSat platform dynamically updates its incident, event, object, or threat detection priorities to meet ever-changing mission needs through uplinked software and configuration driven by users. IQSat offers the industry’s first common payload interface designed to host multiple payloads on a single flexible electrical and mechanical interface, allowing quick and standardized designs to reduce cost, NRE, and time to launch.
XPedite6107 from Extreme Engineering - NXP (formerly Freescale) QorIQ T2081, T1042, or T1022 Processor-Based
Conduction-Cooled XMC/PrPMC Module (four dual-threaded) e6500 cores running at up to 1.8 GHz.
The XPedite6107 provides a compact, cost-effective, rugged computing solution with excellent processing performance per watt. The XPedite6107 supports multiple processor configurations, several I/O options, and up to 8 GB of DDR3 ECC SDRAM. When configured to use the T2081 processor, it maximizes performance with eight virtual
The XPedite6107 provides a compact and cost-effective rugged computing solution with excellent processing performance per watt.
The feature-rich XPedite6107 is available in XMC and PrPMC form factors and supports the NXP (formerly Freescale) QorIQ-based T2081, T1042, or T1022 processors. The T2081 processor provides the same performance as the T2080 processor in a smaller package and is pin-compatible with the T1042 and T1022.
The XPedite6107 supports multiple processor configurations, several I/O options, and up to 8 GB of DDR3 ECC SDRAM. When configured to use the T2081 processor, the XPedite6107 maximizes performance with
ADVERTISERS
eight virtual (four dual-threaded) e6500 cores running up to 1.8 GHz. Each e6500 core implements the NXP AltiVec technology-based SIMD engine, providing DSP-level Floating-Point performance and an extensive inventory of software libraries. The T1042 processor provides a lower-power alternative with four e5500 cores running up to 1.4 GHz. The T1022 provides the lowest power configuration with two e5500 cores running up to 1.4 GHz.
Wind River VxWorks, Linux, and Green Hills INTEGRITY Board Support Packages (BSPs) are also available. Wind River VxWorks and Linux BSPs may optionally be paired with the NXP hypervisor software to facilitate secure partitioning.
COTS Journal (ISSN#1526-4653) is published monthly at; 110 S Rancho Santa Fe, Virgin UT 84779. Periodicals Class postage paid at San Clemente and additional mailing offices. POSTMASTER: Send address changes to COTS Journal, 110 S Rancho Santa Fe PO 790039, Virgin UT 84779.
BEHLMAN LEADS THE PACK AGAIN!
Behlman introduces the first test-proven VPX power supplies developed in alignment with the SOSA Technical Standard. Like all Behlman VPXtra® power supplies, these 3U and 6U COTS DC-to-DC high-power dual output units feature Xtra-reliable design and Xtra-rugged construction to stand up to the rigors of all mission-critical airborne, shipboard, ground and mobile applications.
VPXtra® 1000CD5-IQI
> 6U power module developed in alignment with the SOSA Technical Standard
> Delivers 1050W DC power via two outputs
> VITA 46.11 IPMC for integration with system management
VPXtra® 800D-IQI
> 3U power module developed in alignment with the SOSA Technical Standard
> Delivers 800W DC power via two outputs
> VITA 46.11 IPMC for integration with system management
