AMD’s Ryzen was a huge success for the semiconductor company in 2017, winning new business, reinvigorating the consumer market, and pushing Intel to make the most aggressive realignment of its consumer product families we’d seen in a decade. From the addition of Hyper-Threading to low-end processors to huge price cuts on higher core counts, Ryzen pushed Intel in new directions.
The server market, however, has been a different animal. While AMD has always made it clear that it intended to return to the server space, the company also stated it would do so gradually, ramping up production over time. Now we’re seeing what some of those partnerships will look like, courtesy of a newly announced initiative with Dell.
The three platforms are the Dell PowerEdge R6415, R7415, and the R7425. The R6415 is a 1U configuration, while the R7415 and R7425 are 2U designs (shown below).
An Epyc Forecast
Epyc was a major topic at last week’s conference call. During the call, AMD CEO Lisa Su noted that Sunnyvale’s enterprise, embedded, and semi-custom product family (EESC) sales were up 3 percent that quarter, on the strength of Epyc. She also noted that Epyc had a positive impact on gross margin and AMD’s overall financial position. Asked to give details on what AMD is expected for 2018, Su stated:
I do expect that steady ramp of EPYC as we go through the year. Our target is to be at mid-single-digit unit share by the end of 2018. And so, there would be significant revenue from EPYC as we are in the second half of ‘18.
Dell says AMD’s servers deliver up to 20 percent lower cost of ownership in a single-socket, four-node, VSAN configuration. Those sorts of improvements could appeal to a number of customers, and AMD’s stated goal of hitting a midrange single-digit market share by the end of the year seems plausible (let’s call that 4-6 percent). That relatively small improvement would still constitute a significant increase for AMD, and it’d be in line with year-on-year improvements AMD has seen previously when it fielded competitive server architectures, as shown above.
Nvidia Unveils ‘Grace’ Deep-Learning CPU for Supercomputing Applications
Nvidia is already capitalizing on its ARM acquisition with a massively powerful new CPU-plus-GPU combination that it claims will speed up the training of large machine-learning models by a factor of 10.
Apple Is Working on CPUs With 32 High-Performance Cores: Report
Once the M1 hit a few weeks back, it was clear that the diminutive processor was but a sign of things to come. Reports suggest that Apple will be upping the competitive ante in short order.
Report: Apple Ignored its Partners’ Repeated Violations of Chinese Labor Laws
Apple has reportedly turned a blind eye to repeated violations of Chinese labor law in its partners' factories over the past six years.
FTC Blasts Manufacturer Excuses in Right-to-Repair Report
The FTC has spent the past two years investigating the rights of Americans to repair their own products. It's not happy with the state of things.