AMD launches the most advanced EPYC processor in the world
Aug 26, 2019

AMD has come up with the best processor for data centers in the world. The processor has 64 cores based on the 7-nanometernode and performs at double the speed compared to the previous generation server chips. AMD claims that the latest EPYC processor’s can deliver more than twice the performance for every dollar of Intel’s Xeon CPUs. The quality of high performance for less cost is helping AMD to gain market in the server chip business.
AMD EPYC processor

AMD conveyed that second-generation EPYC CPUs named as Rome-doubles the performance with less power consumption compared to the first generation of server chips named as Naples. The new processors have increased in the performance by 50% at 40% lower cost than Intel’s cascade CPUs. The reduction in the cost, in turn, reduces the cost for cloud and enterprise customers.
Lisa Su, AMD’s CEO said in a statement that its second-generation server chips “set a new standard” for the data center segment. She added that its latest EPYC CPUs are the highest performance x86 processors in the world, taking the throne from rival Intel. Shares of AMD have surged around 17.5% since the announcement of the new EPYC CPUs, which are based on the same Zen 2 architecture as its Ryzen CPUs for PCs.

The Santa Clara, California-based company has been on a competitive edge with Intel. Intel is the world’s largest chip manufacturer and holds more than 95% market share. The market has witnessed the change since last year’s spending surge. Since Su became CEO in 2014, AMD has been in competition with Intel’s Xeon CPUs—the current gold standard in private and public cloud data centers—not only on price but also on performance.

AMD is trying to take advantage of the delays in Intel’s development of server chips with the help of its 10-nanometer node. AMD latest server chips is based on the 7-nanometer node concept. The most advanced and costly chip production technology from TSMC has raised the stakes in AMD’s battle against Intel. AMD using the 7-nanometer node gives AMD’s EPYC CPUs higher core counts than Intel’s Cascade Lake Xeon CPUs, which are based on 14-nanometers.

Intel has plans to release the last generation of server chips in the first half of 2020. The chips are based on 14-nanometers and called Cooper Lake. AMD has pushed out production of server chips based on the 10-nanometer node—which is considered as advanced as the 7-nanometer node. AMD is planning to release the chips in the second half of 2020, giving AMD an opportunity to win over customers from Intel. This will give an edge to AMD over Intel which is struggling to get through its glitches in manufacturing chips.

AMD said sales of its second-generation Rome CPUs are expected to expand “significantly faster” than its first-generation Naples CPUs that were released to the market over a year ago. The company’s server chips have started to gain speed among major cloud vendors. The result is the doubling of AMD’s market share over the last year to 3.4%. The market share mentioned is as during the second quarter of 2019 as per Mercury Research. “Adoption of our new leadership server processors is accelerating,” Su said in a statement.

AMD is aiming to gain double-digit market share in server chips before Intel starts selling its 10-nanometer. The 10-nanometer, Ice Lake CPUs is expected to be released to the market during the second half of 2020. Google said that it had started using AMD’s Rome CPUs in its data centers. Google further plans to offer them to developers over the cloud at cheaper costs than Intel’s Xeon CPUs. AMD is also attracting other major cloud vendors, including Amazon, Alibaba, Baidu, and Microsoft, all of which are big buyers of Intel’s server chips.

“AMD needs to aggressively show that it has a compelling roadmap and step on the gas for the next year to steal as much share as it can,” Patrick Moorhead, principal analyst at Moor Insights & Strategy, said in a research note. Every private enterprise, server manufacturer and cloud vendor “wants more competition in the space to accelerate innovation and lower costs. With that said, none of these customers would adopt AMD if it didn’t have some advantages.”

Moorhead said AMD could win up 10% market share with its new EPYC processors.

The second-generation server chips are based on chiplets. The chiplets connected through AMD’s Infinity Fabric interconnect that boosts the bandwidth between so that they behave as though they share the same die. The new EPYCs integrate up to eight CPUs. The integration is based on the 7-nanometer node from TSMC surrounding the I/O core produced on 14-nanometers from Globalfoundries. The I/O takes care of communications to memory and accelerators such as AMD’s Radeon GPUs.

AMD is building its server processors out of smaller die based on 7-nanometer. The production enables AMD to accommodate more cores at the same power consumption. AMD said the Rome CPUs feature up to 64 cores thereby doubling the core count of its 32-core Naples CPUs. This outclasses Intel’s 56-core Cascade Lake Xeon processors. Intel plans to put up to 56 cores in its upcoming Cooper Lake CPUs. The Cooper Lake CPUs are expected to enter mass production in the first half of 2020.

AMD has another advanced technology to capture market share from Intel’s server processors. The Cascade Lake CPUs support the PCIe Gen 3 standard for connecting to additional storage and accelerators cards for artificial intelligence. AMD’s new EPYC CPU supports 128 lanes of PCIe Gen 4 connectivity, doubling the bandwidth for cloud, enterprise, and high-performance computing customers.

AMD’s latest range of chips ranges from 8 cores to 64 cores. The base frequency of the chips is from 2 GHz to 3.2 GHz. The flagship EPYC 7742 features 64 cores, 128-threads and 256 MB of shared cache and costs $6,950.The base EPYC CPU contains 8 cores, 16 threads and 32 MB of shared cache and costs $450, according to AMD. The memory controller inside every chip can access up to 4 TB of memory at a maximum speed of 3.2 GHz, AMD said.

As per AMD, the top-of-the-line EPYC processor is bound to have 80% better performance for 40% of the price of a high-end Cascade Lake CPU. The Cascade Lake Xeon 8280L based on Intel’s 14-nanometer node contains 28 cores. The chip’s base frequency is 2.25 GHz and power consumption of 225 W. The 64-core CPU gives out 23% more instructions per cycle (IPC) in comparison with 32-core Naples CPU. AMD has also increased the L3 cache from 64 MB to 256 MB.

Intel has been trying to discourage rivals in the data center business by selling a broader portfolio of its products. Intel has started offering its Optane memory to customers who are buying its Cascade Lake CPUs. The Cascade Lake CPUs have been enhanced to handle AI. Intel has also increased on advanced packaging technology that can be used to create more affordable custom chips. The custom chips are made affordable by mixing and matching CPUs, GPUs, FPGAs, ASICs and other chiplets.

Taking on Intel’s dominance in data centers is quite a task. The sales of Intel’s server chips came to $9.9 billion in the first half of 2019. That is quite more than AMD’s total revenue of $6.5 billion in 2018. But Intel has not only lost out on its manufacturing lead to TSMC over the last year but has also started to lose its momentum considerably to AMD. Intel has witnessed its shares losing value by more than 2% in the year 2019.

AMD is still in a position to take huge market share from Intel in the second half of 2019. Profit margins are at 41% in the second quarter, that is up 7% from last year. AMD sees gross margins increasing to 43% in the current quarter. The increase in sales is due to its 7-nanometer Ryzen CPUs, 7-nanometer Radeon GPUs, and 7-nanometer EPYC CPUs. Along with the above AMD aims its 2019 revenue growth to reach 20% excluding sales of semi-custom chips.