Mark Papermaster, CTO and Executive VP of Technology & Engineering at AMD gave us some information regarding EPYC Bergamo, Siena and Genoa-X. For the unaware, Bergamo is the density-optimized version of Genoa aimed at handling massive amounts of data. AMD’s 4th generation EPYC Genoa arrives in 3 flavors namely, Standard Genoa, Bergamo, and Genoa-X.
Bergamo Launching In 1H 2023
AMD unveiled the standard Genoa lineup last month, more on that here. When commenting on Bergamo, Mark Papermaster termed ARM-based solutions as one of AMD’s main competitors. Similarly, the CTO also showed off AMD’s market position by mentioning that Microsoft’s Azure now uses AMD Instinct GPU acceleration for data training.
We’ll talk about Bergamo, our dense core that goes head-to-head with smaller arm cores where you just need throughput processing. Those are all tailored adaptations which we work with hyperscale because we listened because they told us what they needed to have cost-effective solutions and you’ll see more and more accelerators added into that mix. Microsoft announced that they have our instinct, our GPU acceleration now up and running in Azure for their training.
To tackle Graviton and ARM solutions, Bergamo will arrive with a massive 128-core count for tasks that don’t exactly require high clocks. Some workloads benefit from more cores such as Java which does not need the highest frequency but a ton of cores.
Bergamo is still Genoa, it works like Genoa and it executes code similar to Genoa but it is half the size. Zen4c (Bergamo) has smaller cores as compared to Standard Genoa but aims to pack more cores per unit area.
We are adding in first half of this year, what we call Bergamo, which will be with our Zen 4c, we increased staffing to our CPU team, and we added a version of Zen 4. It’s still Zen 4. It runs a code just like Genoa, but it’s half the size.
And that competes head-to-head with Graviton and ARM-based solutions where you don’t need the peak frequency. You’re running workloads like Java workloads, throughput workloads that don’t have to run peak frequency, but you need a lot of cores
Siena Arriving Later Next Year
Based on the SP6 platform, AMD is also planning to launch ‘Siena’ for low-mid ranged server setups. These CPUs will offer up to 64 ‘Zen4’ cores on a lower-cost platform (SP6).
Such a cost-effective solution is aimed at increasing AMD’s market share in sectors such as the telecommunication market. Later in 2023, AMD plans to unveil Siena to consumers.
And then later in 2023, we’re adding the Sienna which is a variant targeted to telecom space. So we’re really, really excited about our TAM growth in server.
Arriving in the same year, Genoa-X is the V-Cache equipped variant of AMD’s Zen4 EPYC CPUs. Just like Bergamo is aimed at density, Genoa-X is aimed at cache-optimized workloads. These CPUs will be equipped with more than 1GB of L3 Cache per socket and will feature upwards of 96 ‘Zen4’ cores based on the 5nm technology.
Keeping things simple and not leaking much, the CTO hinted towards Genoa-X arriving soon. We have the basic definition of AMD’s X3D CPUs along with the targeted market segments.
Well, you mentioned Genoa X. And I didn’t mention that in the variance and I’ll add that now. That’s a version where we stack cash right on top of the CPU and that really is tailored to make high-performance workloads like EDA or database workloads, even more TCO effective.
AMD is planning to go all out in 2023. Right from the first week at CES, we will see budget-oriented Navi32 GPUs along with AMD’s X3D and non-X Ryzen 7000 CPUs. In addition, a slight tease or even a reveal of the Ryzen 7000 Threadripper is on the cards. If that’s not enough, there’s a possibility that AMD will give us a sneak peek at Bergamo and even Genoa-X. Though, the launch is still planned for later next year (Q2 2023 possibly).
Source : seekingalpha