Principal Product Application Engineer

Astera Labs unveils Orion: A compute-agnostic rack-scale architecture. Monday, October 12th 2026 Astera Labs, Inc. (Nasdaq: ALAB), a leader in semiconductor-based connectivity solutions for rack-scale AI infrastructure, today announced Orion: a modular rack-scale reference design with production-grade connectivity that allows customers to choose their own GPU and CPU. To be offered through Astera Labs' ODM and OEM partners, the Orion production-grade reference designs are expected to help enable neocloud and enterprise customers to deploy AI infrastructure faster and run it more efficiently, turning racks built on open standards into token-efficient AI factories. For neoclouds and enterprises to economically serve inference at scale, they must secure sufficient accelerator and CPU capacity while delivering predictable, low-latency performance across increasingly complex AI infrastructure. But because they must commit to that capacity long before they know which accelerator will be available or affordable for their workloads, tying early investment to a single platform carries unnecessary cost and risk. At the same time, ODMs and OEMs are forced to repeatedly solve complex integration, validation, and software challenges for every new accelerator platform - slowing deployments when inference capacity is in critical demand. Orion addresses these challenges with a production-grade reference architecture that decouples rack-scale designs from accelerator and CPU selection. Built on open standards to support industry-leading inference accelerators and CPU ecosystems, Orion is designed to enable operators to deploy infrastructure once and scale inference workloads across the most cost-effective and readily available compute resources - via PCIe plug-in cards - without redesigning racks as technology and market dynamics evolve. At the core of Orion is a high-performance, standards-based, scale-up fabric engineered to connect up to 160 accelerators per pod and deliver predictable, low-latency communication and expanded memory capacity across racks and pods. Orion's optimized PCIe connectivity and composable infrastructure are additionally designed to help accelerators extract more tokens-per-second across AI inference workloads. Orion is delivered as a complete modular rack-scale reference design - including system configurations for the compute tray, fabric tray, memory tray, link board, and compute board - all integrated with COSMOS Rack Manager and COSMOS Fabric Manager orchestration software for management, fleet-wide telemetry, diagnostics, and AI infrastructure intelligence. Pravin Patel, vice president and general manager of the AI Platform Solutions Group at Astera Labs"We're entering a new era of AI inference, where operators are judged on the tokens they deliver per dollar, not the hardware inside their racks. Orion gives AI inference cloud providers a production-grade, compute-agnostic architecture so they can quickly bring inference capacity online, provide outcome-based pricing for their customers, and adapt as the industry and compute landscape evolves." Orion rack reference and system designs are on track for availability through Astera Labs' OEM and ODM partners beginning in Q1 2027. Astera Labs will showcase a powered Orion rack - featuring interchangeable, pre-validated compute trays with different CPU and accelerator configurations at the 2026 Open Compute Project (OCP) Global Summit, taking place October 12-15 in San Jose, California.

Jim Cramer praised Astera Labs but advised investors to avoid the stock due to its high valuation. Speaking on Mad Money on 6 October, he called it "a great company, but one expensive stock" trading at 97 times earnings. Astera reported second-quarter revenue of approximately $392.4 million, up 104% year over year. Its Scorpio fabric switches are expected to become its largest product family in the third quarter, ahead of schedule. The stock trades at approximately 66.4 times forward earnings, compared with 49.3 times for Marvell Technology. Four direct customers accounted for 82% of second-quarter revenue combined. Cramer recommended waiting for a better entry point rather than chasing the recent rally.

Astera Labs unveils PCIe 7 connectivity chips for next-generation AI systems. Astera Labs, Inc. today announced the industry's broadest PCIe 7 signal conditioning portfolio spanning smart retimers, redrivers, cable modules, optical drivers, and optical TIAs. The expanded Aries 7 portfolio offers hyperscalers, neoclouds, OEMs and ODMs a comprehensive connectivity platform that now extends reach from centimeters of PCB trace to 50+ meters over optical cables spanning multiple racks. Multi-rack AI clusters are demanding more bandwidth between every accelerator, memory pool, and networking port. Yet doubling PCIe data rates shrinks signal margins, where a single unstable link can idle costly accelerators that reduce token-per-watt performance. And getting these platforms to production has meant stitching together fragmented multi-vendor solutions with different firmware and software, then re-spinning board designs and validation - resulting in delays in deploying AI infrastructure. Aries has long been the gold standard for reliable PCIe links resulting in high accelerators' utilization. The Aries PCIe 7 expansion carries that same philosophy across every channel a modern cluster needs: use copper where it is the most efficient choice and move to optical where reach demands it - all optimized and monitored through Astera Labs' unified COSMOS software suite. Aries 7 Smart Retimers and Smart Redrivers support the OCP Signal Conditioner Standard Footprint, allowing customers to design once and decide later: as design requirements change, customers can pick different signal conditioners without redesigning the board - optimizing for power and accelerating time-to-market. Aries 7 Smart Optical Drivers and Optical TIAs enable that same intelligence in low power, high density NPO and LPO, designed to deliver optimum performance across multi-rack AI fabrics. Platform partners have put Aries link reliability to the test across multiple generations: AMD and Intel have each validated Aries Smart Retimers in production across their platforms from PCIe 4 through 6. Both companies are now collaborating with Astera Labs to lead the industry's transition to PCIe 7. "The real question for architects isn't retimer vs redriver, or copper vs optical: it's whether they have one comprehensive toolbox with the optionality to pick the right tool for their evolving reach, density and power requirements" said Adit Narasimha, senior vice president, Signal Connectivity Group at Astera Labs. "That is what we built with Aries: the same intelligence and the same software, for the solution the link needs." Aries 7 Smart Retimers, Smart Redrivers, Smart Cable Modules, and Smart Optical Driver and TIAs are expected to begin sampling in Q4 2026 to select hyperscale, OEM, and NPO customers, with broader availability anticipated as platforms complete qualification. Astera Labs will showcase PCIe 7 copper and optical demonstrations next week at the 2026 Open Compute Project (OCP) Global Summit, Oct. 12-15 in San Jose. Shannon, writes, edits and produces Semiconductor Digest's news articles, email newsletters, blogs, webcasts, and social media posts. She holds a bachelor's degree in journalism from Huntington University in Huntington, IN. In addition to her years of freelance business reporting, Shannon has also worked in marketing and public relations in the renewable energy and healthcare industries.