-->

We’re Building the Future of Data Infrastructure

Products
Company
Support

Archive for the ‘Server Processors’ Category

August 31st, 2020

Arm processors in the Data Center

By Raghib Hussain, Chief Strategy Officer and Executive Vice President, Networking and Processors Group

Last week, Marvell announced a change in our strategy for ThunderX, our Arm-based server-class processor product line. I’d like to take the opportunity to put some more context around that announcement, and our future plans in the data center market.

ThunderX is a product line that we started at Cavium, prior to our merger with Marvell in 2018. At Cavium, we had built many generations of successful processors for infrastructure applications, including our Nitrox security processor and OCTEON infrastructure processor. These processors have been deployed in the world’s most demanding data-plane applications such as firewalls, routers, SSL-acceleration, cellular base stations, and Smart NICs. Today, OCTEON is the most scalable and widely deployed multicore processor in the market.

As co-founder of Cavium, I had a strong belief that Arm-based processors also had a role to play in next generation data centers. One size simply doesn’t fit all anymore, so we started the ThunderX product line for the server market. It was a bold move, and we knew it would take significant time and investment to come to fruition. In fact, we have spent six years now building multiple generations of products, developing the ecosystem, the software, and working with customers to qualify systems for production deployment in large data centers. ThunderX2 was the industry’s first Arm-based processor capable of powering dual socket servers that could go toe-to-toe with x86-based solutions, and clearly established the performance credentials for Arm in the server market. We moved the bar higher yet again with ThunderX3, as we discussed at Hot Chips 32.

Today, we see strong ecosystem support and a significant opportunity for Arm-based processors in the data center. But the real market opportunity for server-class Arm processors is in customized solutions, optimized for the use cases at hyperscale data center operators. This should be no surprise, as the power of the Arm architecture has always been in its ability to be integrated into highly optimized designs tailored for specific use cases, and we see hyperscale datacenter applications as no different.

Our rich IP portfolio, decades of processor expertise with Nitrox, OCTEON, and ThunderX, combined with our new custom ASIC capability, and investment in the latest TSMC 5nm process node, puts Marvell in a unique position to address this market opportunity. So to us, this market driven change just makes sense. We look forward to partnering with our customers and helping to deliver highly optimized solutions tailored to their unique needs.

March 17th, 2020

The Next Generation of ThunderX Delivers Performance and Power Advantages to Cloud and HPC Server Markets

By Gopal Hegde, Vice President and General Manager of the Server Processor Business Unit, Marvell

The data centers of today have shifted from a focus on single thread performance to performance at rack scale with performance/watt, performance/$ and overall TCO being the key drivers to deployment. These data centers are making use of servers that are customized for specific workloads. The applications running on these servers are either based on open source software or controlled by the customers who deploy them. Marvell’s ThunderX2® server processor is a leading example of this evolution in the server market with deployments spanning across the cloud and HPC market segments with major customers like Microsoft Azure and the Astra Top 500 Supercomputer installation at Sandia National Laboratories.

We continue to make progress on our roadmap with our third generation Arm® based server processor, ThunderX3™. ThunderX3 is purpose-built for the most demanding workloads in today’s cloud and HPC markets. Utilizing a small die size, ThunderX3 delivers high performance, power efficiency, memory bandwidth and the low memory latency for target cloud and HPC workloads.

Fabricated using the TSMC 7P process, the ThunderX3 processor features up to 96-cores with 4-threads/core delivering a total compute power of 384 threads/socket. The memory interface supports 8-channels of DDR4-3200 with 2 DIMMs/channel. The IO expansion offers 64 lanes of PCIe Gen 4.0 with 16 controllers. The processor supports both single and dual socket configurations. For floating point operations, ThunderX3 features four 128-bit SIMD (Neon) units per core. The device is fully SBSA/SBBR compliant and provides enterprise class RAS and virtualization capabilities. Platforms based on ThunderX3 will sample to customers in mid-year calendar 2020.

Microarchitectural improvements in ThunderX3 provide an overall IPC performance improvement greater than 25% over ThunderX2. This increase along with the increase in the core and DDR frequencies provide an overall gain over 60% in single thread performance versus the previous generation. At the socket level, ThunderX3 provides over 3x higher socket level integer performance and over 5x socket level floating point performance over ThunderX2.

ThunderX3 target markets continue to be specific workloads within the cloud computing and high-performance computing markets where our differentiation delivers strong performance per $ and performance per watt benefits to end customers. In addition, we also enable native Arm workloads such as Android gaming in the cloud. The segments we are focused on constitute approximately 30% of the overall server processor market.

Our target cloud workloads such as big data, databases, media streaming, web tier, elastic search and cloud storage are highly parallel in nature. ThunderX3 features 4 thread simultaneous multithreading (SMT) which enables significant throughput improvements for these workloads. HPC workloads such as EDA and CAE also benefit from multithreading. ThunderX3 is highly power efficient and can maintain higher frequency while operating in floating point-intensive workloads. This capability combined with multiple single instruction, multiple data (SIMD) units and best-in-class memory bandwidth provides a very good performance advantage for target HPC workloads in the areas of quantum physics, quantum chemistry, computational fluid dynamics, genomics, and oil and gas workloads.

The ThunderX3 device is also ideally suited for running native Arm applications that currently deploy on mobile phones and Arm end points as containers or virtual machines in the cloud or at the edge. This enables a variety of emerging use cases such as Android gaming in the cloud, Android and Arm software/application development in the cloud. Marvell has collaborated with NVIDIA in integrating industry-leading GPUs for both HPC and gaming.

Marvell has established a leadership role in the Arm-based server segment with proven performance, large scale production deployments in some of the most demanding environments and a vibrant partner ecosystem that continues to grow.  ThunderX3 is another exciting chapter in this journey and we look forward to sharing more details in the coming months.