Ceva, Inc., has unveiled its fifth generation CEVA-XC DSP architecture, the CEVA-XC20, at MWC 2023. The CEVA-XC20 is based on a new vector multi-threaded massive compute technology that targets next-generation 5G-advanced workloads in a variety of use cases. The new architecture can be used in smartphones, high-end enhanced mobile broadband (eMBB) devices (such as fixed wireless access and industrial terminals) and cellular infrastructure devices (such as base stations, virtualised DU accelerators and beamforming compute in Massive MIMO radios).
The new CEVA-XC20 is based on a vector multi-threaded massive compute technology that is designed to address next-generation 5G-advanced workloads across a broad spectrum of use cases, including smartphones, high-end enhanced mobile broadband (eMBB) devices (e.g. fixed wireless access and industrial terminals) and a range of cellular infrastructure devices (e.g. base stations, virtualised DU accelerators, and beamforming compute in Massive MIMO radios). SoC and ASIC designers incorporating the CEVA-XC20 architecture can avail of its power efficiency to design greener processors that are smaller and lower power, that in turn have a direct positive impact on the environment and society.
Dimitris Mavrakis, senior research director, ABI Research, comments “CEVA’s latest DSP architecture raises the bar for performance and power efficiency in 5G-advanced cellular baseband processing, adopting a unique multi-thread scheme to address the challenging power, performance and area constraints when dealing with complex 5G scenarios. The CEVA-XC20 offers a compelling solution to any wireless semiconductor or OEM developing their own 5G-advanced silicon, and can play a critical role in helping customers to address their sustainability goals in the process.”
The CEVA-XC20 architecture was designed in consultation with CEVA’s Tier 1 OEM customers, with the common aim of improving mobile network performance and power efficiency. The CEVA-XC20 solves the performance challenges posed by next-generation compute-intense 5G-advanced by employing a novel Dynamic Vector Threading (DVT) scheme, which supports true hardware multi-threading, which up until now was only found in general purpose CPU architectures. DVT enables optimal sharing of vector resources between different execution units, resulting in an unprecedented vector utilisation efficiency boost. This technique achieves optimal utilisation of the VLIW architecture and improves core efficiency for common 5G execution kernels, as well as significantly enhancing use cases involving multi-component carriers and multi-execution tasks. This enables increasing the length of the vector processing units, usually consuming the bulk of the area in vector DSPs, while maintaining and even improving the execution efficiency relative to previous generations.
Guy Keshet, vice president and general manager of the mobile broadband business unit at CEVA, states “5G-Advanced and beyond promise ever-increasing cellular bandwidth and lower latency, while being greener and more energy efficient. This creates significant challenges for wireless device companies and mobile network operators who need to deliver on this promise. Our latest CEVA-XC20 DSP architecture addresses these challenges, leveraging more than 30 years of expertise in cellular DSPs at CEVA to deliver incredible power efficiency for the most intense baseband compute use cases. We’re proud to work together with our customers to constantly improve the cellular user experience while reducing the environmental impact of new technology.”
The first core based on the CEVA-XC20 architecture is the CEVA-XC22 DSP, supporting two execution threads using the DVT scheme. The CEVA-XC22 offers a 2.5X improvement in efficiency (performance per watt and area) for essential 5G use cases and computation kernels versus its predecessor. The CEVA-XC22 will also be integrated within CEVA’s holistic baseband platforms, PentaG-RAN for cellular infrastructure and PentaG2-Max for high performance mobile devices, wherein it will power CEVA’s heterogeneous compute platforms, including both DSPs and compute engine accelerators.
Comment on this article below or via Twitter: @VanillaPlus OR @jcvplus
