What Penguin Solutions does

Penguin Solutions, Inc. (NASDAQ: PENG) is a publicly traded AI infrastructure company founded in 1988 and headquartered in Fremont, California. The company operates an AI Factory Platform business model built around five integrated elements: ClusterWareAI software (AI factory operating system for cluster management), MemoryAI (CXL-based KV cache servers and memory modules addressing the AI inference memory wall), OriginAI (validated reference architectures scaling from hundreds to 16,000+ GPU clusters), ComputeAI (advanced computing systems including Altus AMD EPYC, Relion Intel Xeon, and NVIDIA DGX servers), and end-to-end services spanning design, build, deployment, and managed services. The company also retains legacy Stratus fault-tolerant computing platforms (ztC Endurance, ztC Edge, ftServer, everRun), SMART Modular memory products, and Cree LED solutions from prior business lines.

The company serves sovereign AI initiatives (SK Telecom Haein cluster with 1,000+ NVIDIA Blackwell GPUs in Korea), neocloud providers (Voltage Park, Meta RSC since 2017), government and defense (Sandia National Labs Spectra, US Department of Defense), and large enterprises across financial services (Tier-1 institution win Q2 2026), healthcare (private medical school modernization), energy (Shell HPC with immersion cooling), and higher education (Georgia Tech AI Makerspace). Customer relationships are anchored by long-term contracts and managed services engagements.

Revenue is generated predominantly through hardware sales (servers, memory modules, GPU clusters) with growing software and managed services recurring components. Q2 fiscal 2026 net sales were $343 million, with the company raising full-year guidance to 12% growth. Go-to-market combines direct enterprise field sales targeting large accounts, channel partnerships (CDW for mid-market), and strategic alliances with NVIDIA, Dell Technologies, and SK Group. The integrated memory segment grew 63% YoY to 34% of revenue, while the Advanced Computing/AI infrastructure segment faces near-term supply chain constraints. The company employs 1,001-5,000 people across operations in the US, Japan, Korea, India, Malaysia, Singapore, UK, and Taiwan.

Name

Penguin Solutions

Legal name

Penguin Solutions, Inc.

Website

https://penguinsolutions.com

Company type

Public

Founded year

1988

Operating status

Operating

Headcount range

1,001–5,000 employees

Short description

Penguin Solutions is a public AI infrastructure company (NASDAQ: PENG) that designs, builds, and manages AI factory platforms combining software, memory, compute, and services. It serves sovereign AI initiatives, neoclouds, enterprises, and government across financial services, healthcare, energy, and defense.

Ownership category

akta.pro rank

Product category

AI Infrastructure & Data Center Systems

NAICS

Software Publishers (513210), Computer and Peripheral Equipment Manufacturing (33411), Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (51821), Computer Systems Design and Related Services (5415)

SIC

Services-Prepackaged Software (7372), Electronic Computers (3571), Services-Computer Integrated Systems Design (7373)

akta.pro primary industry

End-to-End Enterprise AI Platforms (MLOps & Model Lifecycle Management) (HDAEANAA)

akta.pro secondary industries

AI Compute Virtualization & Scheduling (GPU virtualization, cluster schedulers) (HDAAAAAG), AI Observability, Monitoring & Evaluation Platforms (Drift, Quality, Safety) (HDAEANAF)

Keywords

• AI infrastructure
• High-performance computing
• Data center servers
• Fault-tolerant computing
• Memory modules

Headquarters

HQ city

Fremont

HQ country

United States

HQ region

North America

Offices10 records

Markets served

GTM type

B2B

Offering type

Hardware or Manufacturing

Cost components

Supply Chain, Personnel, Operations, Technology or R&D, Marketing or Sales, Infrastructure

Revenue model

1. Integrated Memory: Memory products including DRAM modules, CXL solutions, flash storage, and SMART Modular products. Grew 63% year-over-year in Q2 2026, accounting for 34% of total revenue. Strong growth driven by AI-driven DRAM demand and DDR5 upgrade cycle.
2. Advanced Computing / AI Infrastructure: AI factory platform solutions including servers, GPU clusters, storage systems, and OriginAI validated designs. Previously called Embedded Computing segment. Faces near-term headwinds from supply chain constraints and slower Advanced Computing conversion.
3. Professional and Managed Services: End-to-end services including design, build, deployment, and managed services for AI infrastructure. NVIDIA DGX-Ready Managed Services provider. Services help customers deploy, optimize, and scale AI environments.
4. Fault Tolerant Computing Platforms: Stratus platforms including ztC Endurance, ztC Edge, ftServer, and everRun for continuous availability applications in critical infrastructure.
5. ClusterWareAI Software: Software platform operating system for AI factory management. Recently expanded with AI Factory Operations Agent featuring conversational natural language interface for GPU cluster performance insights and automated GPU remediation for Kubernetes workloads.

Go-to-market motion3 records

Distribution channels5 records

Marketing channels7 records

Core offering

Penguin Solutions designs, builds, deploys, and manages full-stack AI factory infrastructure combining proprietary software (ClusterWareAI), memory solutions (MemoryAI KV Cache Servers, SMART Modular DDR5/CXL modules), validated reference architectures (OriginAI), compute systems (Altus AMD EPYC, Relion Intel Xeon, NVIDIA DGX, Dell servers), and fault-tolerant platforms (Stratus ztC). It sells AI infrastructure hardware, software, and end-to-end services to enterprises, governments, sovereign AI initiatives, and neocloud providers.

Product overview

Penguin Solutions operates as an AI Factory Platform Company offering a comprehensive full-stack AI infrastructure portfolio. The platform combines five core elements: ClusterWareAI (AI factory platform operating system software providing cluster management and AI-driven operations), MemoryAI (CXL-based KV cache servers and memory solutions for AI inference), OriginAI (validated reference designs scaling to 16,000+ GPU clusters), ComputeAI (advanced computing systems), and end-to-end services (design, build, deployment, and managed services). The company also offers fault-tolerant computing platforms (Stratus ztC Edge, ztC Endurance, ftServer, everRun), integrated memory products under SMART Modular brand (DDR5 modules, CXL solutions, Zefr ZDIMMs, SSDs), LED solutions under Cree LED brand, and GPU/compute hardware including Altus AMD EPYC servers, NVIDIA DGX systems, and Dell AI-optimized infrastructure. The portfolio enables enterprises to design, build, deploy, and manage AI workloads from training to inference at scale.

Differentiator

Problem solved

Functional benefit

Brands

• OriginAI: AI factory infrastructure solution built on pre-defined AI architectures that can scale from hundreds to over 16,000 GPU clusters
• ClusterWareAI
• MemoryAI
• ComputeAI
• SMART Modular
• Cree LED
• SMARTsemi
• Stratus

Products and services

• ClusterWareAI AI factory platform operating system software that unifies and automates cluster deployment and management. Transforms bare-metal hardware, network, and software resources into high-performance cluster environments with zero-touch provisioning, real-time monitoring, and AI-driven operations. Sold to enterprises operating GPU clusters.
• MemoryAI KV Cache Server Industry's first production-ready CXL-based KV cache server delivering up to 11TB of memory capacity. Offloads KV pairs from GPU memory to solve memory wall challenges in AI inference, enabling higher GPU utilization and reduced time-to-first-token. Sold to enterprises running AI inference at scale.
• OriginAI Infrastructure Validated AI factory infrastructure solution built on pre-defined AI architectures scaling from hundreds to over 16,000 GPU clusters. Integrates validated technologies with Penguin's cluster management software and expert services for designing, building, deploying, and managing AI infrastructure at scale. Sold to enterprises and sovereign AI initiatives.
• ComputeAI Advanced computing systems and infrastructure elements optimized for AI workloads, supporting AI training, inference, and data-intensive applications. Built for scalability and performance. Sold to enterprises deploying AI infrastructure.
• Altus AMD EPYC Servers High-performance servers powered by AMD EPYC processors optimized for AI and HPC workloads. Available with immersion cooling technology for energy-efficient computing. Sold to enterprises and HPC customers.
• Relion Intel Xeon Servers Enterprise servers powered by Intel Xeon processors for AI infrastructure, HPC, and data-intensive applications. Sold to enterprises and HPC customers.
• GPU Accelerated Servers Servers designed for GPU-accelerated computing workloads, supporting enterprise AI training and inference deployments. Sold to enterprises deploying GPU clusters.
• NVIDIA DGX Systems NVIDIA DGX AI supercomputing systems integrated and managed by Penguin Solutions as an NVIDIA DGX-Ready Managed Services partner. Provides full-stack AI infrastructure for training and inference workloads. Sold to enterprises and sovereign AI customers.
• Dell AI Optimized Hardware Dell PowerEdge servers and PowerScale storage integrated into Penguin's AI factory solutions. Featured in Deepgram voice AI deployment with NVIDIA RTX PRO 6000 GPUs. Sold to enterprise AI customers.
• Stratus ztC Edge Secure, rugged, highly automated edge computing platform with self-protecting and self-monitoring features that reduce unplanned downtime and ensure continuous availability of business-critical applications at the network edge. Sold to critical infrastructure operators.
• Stratus ztC Endurance Intelligent, predictive fault-tolerant computing platform enabling 99.99999% compute platform availability. Combines built-in fault tolerance, proactive health monitoring, and serviceability for mission-critical deployments. Sold to critical infrastructure operators.
• Stratus ftServer Fault-tolerant server platform designed for continuous availability in enterprise data center environments. Sold to critical infrastructure and enterprise customers.
• Stratus everRun Software solution that pairs two servers via virtualization to create protected and replicated virtual machines, ensuring applications run without interruption or data loss. Sold to enterprise customers needing continuous availability.
• Stratus V Series Virtualization-ready fault-tolerant computing platform for enterprise applications requiring high availability. Sold to enterprise customers.
• SMART Modular CXL-Based Memory Solutions Compute Express Link (CXL) memory expansion solutions including add-in-cards and memory modules for AI/ML workloads, data centers, and HPC environments requiring high-bandwidth memory expansion. Sold to enterprise, hyperscaler, and AI infrastructure customers.
• SMART Modular DDR5 Memory Modules DDR5 DRAM modules including standard, ECC, and high-capacity configurations for enterprise servers and data centers. Includes 64GB DDR5-6400 ECC CSODIMM for harsh environment deployments. Sold to enterprise and industrial customers.
• SMART Modular Zefr ZDIMMs Ultra-high reliability ZDIMM memory modules ideal for data centers, hyperscalers, and HPC platforms running large memory applications requiring maximum compute availability. Sold to hyperscalers and HPC customers.
• SMART Modular Rugged/Industrial Memory Memory modules designed for rugged and industrial environments with extended temperature ranges and enhanced reliability for demanding conditions. Sold to industrial, telecom, and edge computing customers.
• SMART Modular Value Memory Cost-effective memory solutions for general enterprise server and storage applications. Sold to enterprise server and storage customers.
• SMART Modular Data Center SSDs Next-generation solid-state drives designed for hyperscaler, hyper-converged, enterprise, and edge data centers meeting stringent storage demands. Sold to hyperscalers, enterprises, and data center customers.
• SMART Modular Embedded SSDs Embedded solid-state drives for industrial and embedded computing applications requiring reliable flash storage. Sold to industrial and embedded computing customers.
• SMART Modular RUGGED SSDs Ruggedized solid-state drives designed for harsh environment storage applications. Sold to defense, industrial, and harsh-environment customers.
• Cree LED XLamp XE-B LEDs High-intensity LEDs in compact 0.9x1.4mm package delivering up to 60% higher intensity than standard 1x1mm LEDs for indoor directional, architectural, entertainment, and aftermarket automotive lighting. Sold to lighting manufacturers.
• Cree LED OptiLamp LEDs Display technology integrating driver and control intelligence directly into every LED pixel, eliminating external driver ICs. Delivers enhanced brightness, 24-bit control per channel, and power efficiency improvements. Sold to display and lighting manufacturers.
• Cree LED L2 PCBA Modules Level 2 fully assembled LED PC board assemblies in standard and custom configurations for lighting manufacturers, providing supply stability and flexible production. Sold to lighting manufacturers and OEMs.
• AI Infrastructure Services Comprehensive services including design, build, deployment, and managed services for AI factory infrastructure. Certified NVIDIA DGX Managed Services provider offering end-to-end AI lifecycle support. Sold to enterprises and sovereign AI customers.
• Cluster Integrity Assessment Assessment service evaluating existing GPU cluster health, performance, and optimization opportunities. Sold to enterprises operating GPU clusters.
• SMART Memory Test Labs

Quantifiable outcome

• 63% year-over-year growth in integrated memory segment
• +5 more outcomes

Named customers10 records

Segments8 records

Ideal customer profiles5 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

AI capability10 records

Feature6 records

Partnerships

15 partnerships are on record, tiered core, minor and secondary.

• NVIDIAcoreTechnology or Integration · 23 June 2026NVIDIA AI Factory Specialized Partner within NVIDIA Partner Network. DGX-Ready Managed Services Provider. Decade-long collaboration delivering AI factories for hyperscalers and enterprises. Joint work on Haein AI Factory in Korea.
• Dell TechnologiescoreGTM or Marketing Partner · 3 June 2026Global Alliances Americas AI Partner of the Year 2026. Collaboration on Full-Stack AI Factory Platforms combining Penguin software, memory, compute with Dell infrastructure. Deepgram partnership leverages Dell PowerEdge servers and PowerScale storage.
• SORBA.aiminorStrategic or Co-development Partner · 8 April 2026Strategic partnership to deliver fault-tolerant, on-prem industrial AI solutions. Enables real-time machine learning, autonomous control, and secure data management for industrial environments.
• DeepgramcoreStrategic or Co-development Partner · 17 March 2026Strategic collaboration to deploy optimized AI inference infrastructure for enterprise voice AI. Solution uses Dell PowerEdge servers with NVIDIA RTX PRO 6000 Blackwell GPUs and Dell PowerScale storage for low-latency, high-concurrent speech-to-text, text-to-speech, and voice agent applications.
• CDWsecondaryChannel Partner/ Reseller/ Distributor · 6 May 2025Agreement signed May 2025 expanding customer reach for Penguin Solutions' AI infrastructure offerings. CDW provides access to broader mid-market customer base in North America.
• RebellionsminorStrategic or Co-development Partner · 4 March 2025Strategic collaboration initiative to advance global AI data center ecosystem. South Korean AI chip company Rebellions partnering with Penguin and SK Telecom.
• SK TelecomcoreStrategic or Co-development Partner · 9 January 2025SK Telecom and Penguin Solutions collaborated on Haein, one of Korea's largest sovereign AI GPU-as-a-Service clusters with over 1,000 NVIDIA B200 GPUs. SK Telecom made $200M strategic investment in Penguin Solutions (July 2024). Collaboration agreement signed January 2025 for next-generation AI data center solutions.
• Voltage ParksecondaryImplementation/ SI/ Consulting Partner · 11 July 2024Penguin Solutions selected as managed services partner for Voltage Park's NVIDIA GPU clusters. Voltage Park CEO praised Penguin's end-to-end ability to deliver, optimize, and support complete multi-tenant environment.
• SK HynixcoreStrategic or Co-development PartnerSK Hynix, SK Telecom, and Penguin Solutions formed strategic collaboration for AI data center ecosystem development. SK Hynix provides memory expertise and capital support. SK Group subsidiary AI Company acquiring Penguin shares from SK Telecom as part of AI asset consolidation.
• AMDsecondaryTechnology or IntegrationAMD EPYC processors power Altus servers and MemoryAI KV Cache Server. Collaboration with Shell on HPC cluster using AMD EPYC 9654 processors with immersion cooling. Longstanding CPU partnership.
• ShellsecondaryImplementation/ SI/ Consulting PartnerShell uses Penguin Solutions' Altus AMD EPYC servers with immersion cooling technology for HPC cluster upgrade in Houston data center. 864 dual-socket systems with 165,888 cores drawing from 100% renewable power. Supports Shell's net-zero emissions goal by 2050.
• MetacoreStrategic or Co-development PartnerFive-year partnership providing AI-optimized architecture and managed services for Meta's AI Research SuperCluster (RSC). Deployed 16,000 GPUs. Built on 2017 partnership that established Meta's first-generation AI infrastructure. Collaboration continues for next-generation AI systems.
• NextSiliconsecondaryStrategic or Co-development PartnerPenguin Solutions deploys NextSilicon's Maverick-2 runtime-reconfigurable accelerators as part of Sandia National Laboratories' Vanguard program. Spectra supercomputer passed acceptance testing in 2026. NextSilicon claims 10x performance and 60% power efficiency vs NVIDIA GPUs.
• Pure StoragesecondaryTechnology or IntegrationPenguin Solutions supports Pure Storage FlashBlade//EXA introduction. Integration partnership for AI and HPC storage solutions.
• IonQminorStrategic or Co-development PartnerCollaboration with SK Telecom and IonQ for sovereign AI ecosystem development. Quantum computing partnership for advanced AI capabilities.

Scale indicators12 records

Recent moves8 records

Expansion highlights6 records

Direct peers

• SuperMicro: SuperMicro is a direct competitor in GPU-accelerated servers and AI infrastructure, offering similar rack-scale systems with NVIDIA and AMD processors. Both companies target the same enterprise and neocloud customers building large-scale AI clusters.
• Hewlett Packard Enterprise (HPE): HPE competes directly with Penguin Solutions in HPC and AI server systems through its Cray HPC portfolio and ProLiant/Apollo AI servers. Both target sovereign AI, enterprise, and research computing buyers with full-stack hardware-plus-services offerings.
• Dell Technologies: Dell is both a strategic partner (Americas AI Partner of the Year 2026) and a competitor, selling PowerEdge servers with NVIDIA GPUs and its own AI factory software stack. Customers can buy similar AI infrastructure directly from Dell, making Dell the most overlap-heavy peer in Penguin's portfolio.
• Lenovo: Lenovo competes in AI/HPC server infrastructure with ThinkSystem and ThinkEdge products integrated with NVIDIA GPUs. Both companies target enterprise, neocloud, and government customers deploying GPU clusters for AI training and inference.
• Lambda: Lambda is a direct competitor in AI infrastructure, offering GPU clusters, cloud services, and AI workstations to enterprises and AI labs. Both Penguin Solutions and Lambda target customers building production AI workloads at scale.

Broad incumbents

• IBM: IBM is a broad incumbent competing in HPC and enterprise AI infrastructure via Power Systems, watsonx, and consulting services. While IBM's portfolio is much wider, it overlaps with Penguin in HPC for government, research, and large enterprise customers.

Emerging players

• CoreWeave: CoreWeave is an emerging AI cloud provider built on NVIDIA GPU infrastructure. While primarily a buyer of GPU systems (rather than a systems integrator), CoreWeave represents an alternative deployment path for enterprise AI workloads, partially overlapping Penguin's neocloud and enterprise AI infrastructure segments.
• Nebius: Nebius is an emerging AI infrastructure provider offering GPU cloud services built on NVIDIA hardware. Similar to CoreWeave, it represents an alternative AI compute deployment model that intersects with Penguin Solutions' enterprise and neocloud customer segments.
• Crusoe: Crusoe builds and operates large-scale AI GPU clusters, partly for its own cloud and partly for customers. It competes for AI infrastructure deployment mandates and represents an alternative path for enterprises seeking GPU capacity at scale.
• GMI Cloud: GMI Cloud is an emerging AI infrastructure provider focused on NVIDIA GPU clusters for enterprise and AI lab customers. It represents a smaller but growing alternative in the same AI factory/neocloud category where Penguin Solutions competes.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat5 records

Key risks6 records

Key highlights7 records

Customer concentration

• Xx.com/weare_penguin
• YouTubeyoutube.com/channel/UCGF5cVokG9riAs_8R2C6kmw
• Facebookfacebook.com/Penguin-Solutions-107438131880011
• LinkedInlinkedin.com/company/penguinsolutions

Compliance3 records

Revenue estimate

Valuation estimate

Number of profiles

Profiles12 records

Subsidiaries3 records

Funding overview

Funding rounds3 records

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A2 records

Investments3 records

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Penguin Solutions