Reports
Published market reports covering AI chips, data-center compute, storage infrastructure, products, technology, and forecasts.
Ultra-low power AI chips for always-on applications
This report covers ultra-low power AI chipsets under 100 mW used in always-on applications such as smart speakers, wearables, wake-word listening, sensor-signal analysis, and people monitoring, with quantitative data for 2021–2027 on market size, unit volume, and ASP by architecture (CPU, GPU, FPGA, ASIC), application type, and geography (North America, Europe, Asia Pacific, Latin America, and Middle East & Africa), plus vendor and product listings, IP providers, sensors, speech and vision pipelines, software stacks, benchmarks, start-ups, and ecosystem players in a segment already shipping tens of millions of units.
AI Chips Technology and Market Overview: CPUs, GPUs, FPGAs, and ASICs for Training and Inference
This report includes AI chipset market data for CPUs, GPUs, FPGAs, ASICs, and SoC accelerators used in training and inference, with 2020–2027 world-market revenue, year-on-year growth, estimated AI workloads on enterprise GPUs and CPUs, and splits by architecture, training versus inference, power class (high above 50 W, medium 5–50 W, low 0.1–5 W, and ultra-low below 0.1 W), enterprise versus edge, and geography (North America, Europe, Asia Pacific, Latin America, and Middle East & Africa), plus neural-network compute needs, product details, start-up and M&A activity, and a company listing across data-center (50+ W), mid-power (5–50 W), low-power (0.1–5 W), and always-on ultra-low-power segments.
AI driven storage Infrastructure: Markets, technology, products and market forecast
This report includes AI-driven storage market data for 2020–2027 covering revenue by product category, capacity by region, and unit shipments, ASP, and revenue for flash arrays, direct-attached storage, and scale-out storage, with geographic splits and a training-versus-inference cut, plus data-type and capacity requirements, object storage and databases, flash-array interfaces and scale-up versus scale-out architectures, vendor product comparisons with publicly identified buyers, start-up assessments, and vertical-market use cases that generate large AI data volumes.
Data center AI Compute Infrastructure: Markets, technology, products and market forecast
This report includes data-center AI compute infrastructure data for 2020–2027 covering revenue by product category and region, plus unit shipments, ASP, and revenue for accelerator cards, workstations, and servers, with splits by geography, training versus inference, and server buyers (hyperscalers versus enterprises), along with AI usage and infrastructure purchase across 25 enterprise verticals, CPU/GPU/FPGA/ASIC acceleration, bus interfaces, scaling and utilization, storage and networking attachments, vendor product assessments with publicly identified buyers, start-up coverage, and company profiles.
AI chipsets for data center: Markets, technology, products and market forecast
This report includes enterprise and data-center AI acceleration chipset data through 2027 covering world-market revenue by category and region, plus unit shipments, ASP, and revenue for AI CPUs, accelerator servers, and workstations, with a server-revenue split by buyer type, along with CPU, GPU, FPGA, and ASIC product details for training and inference, hyperscaler versus enterprise demand, use cases and data types, architecture and bus-interface characteristics, vendor assessments with publicly identified buyers, start-up coverage, and company profiles.
