Data Science Intern

Marvell data-center revenue rose 46% year over year in Q2. Will Chu said Marvell chose around 2018-19 to skip 7nm and move from 12nm and 16nm products directly to 5nm. Published Oct 26, 2026, 5:53 PM EDT Marvell's data-center revenue rose 46% year over year in its second quarter, driven by AI-related demand for electro-optics, custom products, storage and switching. The segment's growth rate exceeded the company's overall revenue increase. Marvell posted net revenue of $2.739 billion for the three months ended Aug. 1, 2026, up 37% from a year earlier. The 46% figure applies to the data-center business, while the 37% figure covers the company's total net revenue. Marvell did not quantify how much, if any, of the segment's growth came from the earlier move to 5nm products. Will Chu, Marvell's executive vice president and general manager of Custom Cloud Solutions, said the company made that process-node choice around 2018-19. Chu said Marvell moved directly from 12nm and 16nm products to 5nm, skipping 7nm. "We needed to be a leader and not a follower," Chu said. Marvell has executed multiple 5nm designs, is progressing through 3nm designs and is developing a 2nm platform. The company also demonstrated 2nm optical technology at ECOC, including 400G-per-lane optical PAM4. Additional demonstrations included 800G ZR/ZR+ with MACsec and 1.6T ZR and coherent-lite technology. These were demonstrations, not announcements of commercial shipments. Marvell Chairman and CEO Matt Murphy said, "AI-related bookings remain exceptionally robust." The bookings comment reflects demand, while the company's 46% figure measures year-over-year data-center revenue growth. Neither statement quantifies a revenue contribution from the 2018-19 decision Chu described. John Kim. John Kim reports on the digital-asset business for TokenPost. Send corrections or tips to [email protected].

Marvell Technology has signed a $3 billion revolving credit agreement, doubling the $1.5 billion facility it replaces. Bank of America, N.A. is the administrative agent, and the company disclosed the deal in an 8-K filing on October 23, 2026. The agreement took effect on October 22, 2026, and no loans were outstanding on that date. Marvell may raise commitments by up to $2 billion more, subject to conditions. The facility matures after five years, with an option to request a one-year extension. Proceeds may fund working capital and general corporate purposes. Borrowings carry interest at SOFR-based or alternate base rates, plus a margin set by Marvell's debt ratings. The company must keep its leverage ratio at or below 4.00 to 1.00, rising to 4.50 to 1.00 for qualifying acquisitions of at least $500 million.

BofA crowns Nvidia & AI chip titans as $1 trillion cloud spending looms: A deep dive into the future of AI infrastructure. Editorial Team Crypto Analyst Oct 22, 2026 In a landscape increasingly dominated by the relentless march of artificial intelligence, Bank of America (BofA) has cast a discerning eye over the semiconductor and technology sectors, identifying a select group of companies poised to become the ultimate beneficiaries of an impending colossal wave of cloud infrastructure spending. Despite a recent market wobble that saw these very stocks experience notable dips, BofA remains steadfast in its conviction, naming Nvidia, Intel, Marvell Technology, Micron, Lam Research, and Synopsys as top contenders in the race to power the AI revolution. This bullish outlook, championed by BofA analyst Vivek Arya, underscores a foundational shift in enterprise and cloud spending, anticipating a staggering $1 trillion investment in cloud infrastructure that will inevitably fuel demand for cutting-edge AI hardware and software. The recent market downturn, where Marvell dropped 8.29%, Intel 6.44%, Micron 4.78%, Synopsys 3.99%, and Nvidia 3.93%, might appear paradoxical against such a strong long-term forecast. However, for a seasoned analyst, such short-term corrections often represent profit-taking or broader market recalibrations rather than a fundamental questioning of underlying growth trajectories. The core thesis remains robust: the explosion of AI, particularly the widespread adoption of open-weight AI models, demands an unprecedented level of computational power, data storage, and specialized infrastructure. This demand is not fleeting but represents a multi-year investment cycle that will redefine technological capabilities across industries. The unstoppable surge of AI infrastructure. The impetus behind BofA's optimistic forecast is the exponential growth in demand for AI processing capabilities. As AI models become increasingly complex, featuring billions, even trillions, of parameters, the resources required to train, deploy, and run them skyrocket. From natural language processing to computer vision, generative AI, and advanced analytics, every facet of modern computing is becoming AI-centric. Hyperscale cloud providers - the likes of Amazon Web Services, Microsoft Azure, and Google Cloud - are at the forefront of this investment, building out vast data centers equipped with the specialized silicon needed to handle these workloads. The spread of open-weight AI models, as highlighted by Vivek Arya, further democratizes access to AI development, accelerating innovation and consequently driving greater demand for the underlying infrastructure from a broader developer base. This insatiable demand extends beyond just processing power. It encompasses every layer of the AI stack: high-bandwidth memory for rapid data access, advanced networking solutions to ensure seamless data flow, and sophisticated design and manufacturing tools to create these marvels of silicon. The $1 trillion cloud spending forecast is not just a projection; it's a reflection of the critical need for enterprises globally to integrate AI into their operations, a journey that begins with robust, scalable infrastructure. Nvidia: the undisputed AI kingpin. At the apex of this AI hardware revolution sits Nvidia. While it experienced a dip alongside its peers, Nvidia's position as the dominant force in AI acceleration remains virtually unchallenged. Its Graphics Processing Units (GPUs), particularly the Hopper and Blackwell architectures, are the de facto standard for AI training and inference. Beyond hardware, Nvidia's CUDA platform provides a comprehensive software ecosystem that has fostered a vibrant developer community and a vast library of AI applications, creating a powerful moat around its technology. As AI workloads become more complex and widespread, Nvidia's full-stack approach, integrating hardware, software, and networking, positions it as an indispensable partner for any entity serious about AI. The ensemble cast: powering AI from silicon to software. While Nvidia captures significant headlines, the success of the AI era is a collaborative effort, relying on a sophisticated supply chain. BofA's list of beneficiaries reflects this intricate ecosystem: * Intel: Though historically focused on CPUs, Intel is aggressively pursuing the AI accelerator market with its Gaudi series and foundational CPU contributions to data centers, crucial for general-purpose computing that complements GPU-intensive tasks. * Marvell Technology: A leader in data infrastructure semiconductor solutions, Marvell's custom silicon and networking chips are vital for handling the massive data flows within AI-powered data centers, ensuring efficient communication between GPUs, CPUs, and memory. * Micron: As a primary provider of memory solutions, particularly High Bandwidth Memory (HBM), Micron is critical. AI models are memory-hungry, requiring fast and abundant memory to feed the powerful GPUs, making Micron an indispensable component of the AI infrastructure buildout. * Lam Research: This company is a titan in the semiconductor equipment industry, providing the manufacturing tools that enable chipmakers to produce the advanced chips needed for AI. As chip geometries shrink and complexity grows, Lam Research's role in fabrication becomes even more crucial. * Synopsys: A leader in Electronic Design Automation (EDA) software and semiconductor IP, Synopsys is foundational. It provides the tools and intellectual property that chip designers use to create the incredibly complex AI chips, ensuring their functionality, performance, and manufacturability. These companies collectively form the backbone of the AI value chain, from design to manufacturing to deployment. Navigating market volatility: A long-term perspective. The recent dip in these AI-centric stocks, as observed by BofA, serves as a crucial reminder of market dynamics. Such pullbacks can be attributed to various factors: broader market corrections influenced by macroeconomic concerns like interest rate fluctuations, profit-taking after significant rallies, or even sector rotation as investors rebalance portfolios. However, as a senior analyst observing these trends from a macro and technological perspective, it is imperative to distinguish short-term market noise from long-term fundamental shifts. The underlying demand for AI infrastructure is not diminishing; it is accelerating. The $1 trillion cloud spending forecast represents a structural, secular trend, not a transient fad. Investors with a long-term horizon who understand the foundational role these companies play in the global AI buildout may view such dips as potential entry points rather than red flags. From a broader financial technology standpoint, the implications of this massive AI infrastructure investment are profound. The enhanced computational capabilities will not only drive innovations in traditional finance, such as algorithmic trading, fraud detection, and personalized banking, but also underpin the development of more sophisticated Web3 applications, decentralized AI networks, and advanced blockchain analytics. The robustness and scalability of the underlying AI hardware ecosystem are direct enablers of the next generation of digital finance and decentralized innovation, highlighting the interconnectedness of seemingly disparate tech sectors. Conclusion. Bank of America's assessment firmly establishes Nvidia and its semiconductor peers as the architects of the AI future. The confluence of surging AI model complexity, the widespread adoption of open-weight models, and an impending $1 trillion investment in cloud infrastructure creates an undeniable tailwind for these companies. While market volatility will always present short-term fluctuations, the long-term trajectory for these AI infrastructure providers remains exceptionally strong. As the global economy continues its inexorable shift towards AI-driven paradigms, the companies building the foundational silicon and software will not merely benefit; they will define the technological landscape for decades to come, shaping everything from enterprise efficiency to the future of digital assets and decentralized intelligence. #Crypto #MarketUpdate #Blockchain