• Skip to main content
  • Skip to search
  • Skip to footer
Cadence Home
  • This search text may be transcribed, used, stored, or accessed by our third-party service providers per our Cookie Policy and Privacy Policy.

  1. Blogs
  2. Breakfast Bytes
  3. Rapid Adoption Kits for Arm's Premium Mobile Platforms
Paul McLellan
Paul McLellan

Community Member

Blog Activity
Options
  • Subscribe by email
  • More
  • Cancel
Rapid Adoption Kit
RAK
digital full flow
mobile
verification full flow
ARM

Rapid Adoption Kits for Arm's Premium Mobile Platforms

25 May 2021 • 4 minute read

 breakfast bytes logoarm total compute graphic for cadence raksToday, Arm announced its new lineup of processors for mobile. These are the first Arm v9 instruction set (ISA) CPUs, new Mali GPUs, and new system interconnect. Cadence announced rapid adoption kits for the premium platform, which consists of the Cortex-X2, Cortex-A710, and Cortex-A510 CPUs, and the Mali-G710 GPU.

Almost exactly a month ago, Arm announced its new lineup of processor IP for data centers. Cadence announced the availability of Rapid Adoption Kits (RAKs) for those processors the same morning. You can read about that in my post Rapid Adoption of the Arm Server-Class Processors.

Today, it is the turn of mobile. As part of our collaboration with Arm, Cadence has optimized its digital and verification full flows on 5nm and 7nm process technologies to drive adoption of the new Arm premium mobile platform. In addition, Cadence delivered 5nm and 7nm RTL-to-GDS digital flow RAKs to enable customers to achieve power, performance, and area (PPA) goals and accelerate time to tapeout.

Often, I write about digital full flow or RAKs, and I just assume everyone knows the names of all the Cadence tools that go into it, just like I do. But obviously that is not always the case. The tools optimized for the RAK are:

  • cadence digital full flowGenus Synthesis Solution
  • Modus Test
  • Innovus Implementation System
  • Quantus Extraction Solution
  • Tempus Timing Signoff Solution and ECO Option
  • Voltus IC Power Integrity Solution
  • Conformal Equivalence Checking and Conformal Low Power 

For advanced mobile designs using these new Arm processors, there are many features in the flow that are optimized for these advanced designs in advanced process nodes.For example, the Cadence iSpatial technology provides an integrated, predictable implementation flow so users can achieve faster design closure. The flow also incorporates an innovative hierarchical technology to deliver optimal turnaround time on large high-performance CPUs. Innovus's GigaOpt power optimization capability significantly reduces dynamic power consumption (very important in mobile, of course). And the Tempus ECO Option provides signoff-accurate final design closure using path-based optimization (PBO), which enables users to achieve PPA targets quickly.

Chin-Chi on Moore's Law and 3nm

Chin-Chin Teng, SVP of our digital full flow (that's not his official title!), has an opinion piece Moore's Law Is Still Accelerating. It takes a look at 3nm, the next node after 7nm and 5nm (the nodes targeted by the RAKs being announced today). He starts:

There has been much said about the slowing, or even ending, of Moore’s Law. But I'm here to say that Moore’s Law is still accelerating. The semiconductor industry isn’t giving up so easily, and the pace of innovation is faster than ever. The concept of More than Moore or redefining the goal from chip-specific gains to system-specific gains is also legitimate. Semiconductor foundries and EDA partners have made tremendous strides in the past five years, speeding innovation, building fabs and improving yield faster than ever before.

So how can I justify my claim? I’m looking at Moore’s Law differently, measuring the pace of new node arrivals. Let’s look at the major nodes: 28nm, 16nm, 7nm, and now 3nm. We can see that the node-to-node transitions have accelerated from a three-year pace (from 28nm to 16nm) to a one-and-a-half to two-year pace (from 7nm to 5nm). And the transition from 5nm to 3nm occurred faster than ever as we now see 3nm rapidly approaching readiness.

Read the whole piece.

Arm-Based Instances at Amazon AWS

graphs showing aws graviton2 expanding in ec2

No matter how wonderful the specs, the success of a high-tech product is measured in business success. We can all think of products that were wonderful from a technical point of view but never achieved traction. Often, despite the paranoia about being late to market, some of these products failed simply by being too early. For more on that, see my post It's the Second Mouse That Gets the Cheese. AWS's Arm-based Graviton2 is expanding its share in AWS's data centers. It now comprises 14% of AWS EC2 instances (note that the graph starts at 50% so is a bit misleading). Vendor A (I think we all know who "Vendor A" is) still has about 70% share in the installed instances. But perhaps the more dramatic number is shown in the pie chart on the right—almost half of the new EC2 instances added during 2020 were Graviton2-based. This is actually pretty surprising. Graviton2 was announced at the Re:invent conference at the end of 2019, and so from a standing start has risen to half (almost) of new installations. If you assume any sort of ramp during 2020, then presumably Graviton2 exited the year with a run rate of well over half of new EC2 instances being installed.

Just to avoid any confusion, Graviton2 is based on Arm Neoverse-N1, not one of the new server processors announced last month. There hasn't been enough time for that. But I'm not exactly going out on a limb when I say that I'm sure that there will be new Graviton SoCs based on the N2 or the V1 (or perhaps both) in AWS/Annapurna's future.

 

Sign up for Sunday Brunch, the weekly Breakfast Bytes email.