AWS puts latest homebrew Graviton3 Arm processor in production

Just one instance type for now, cheaper than third-gen Xeons and Epycs


Amazon Web Services has made its latest homebrew processor, the Graviton3, available to rent in its Elastic Compute Cloud (EC2) infrastructure-as-a-service offering.

The cloud colossus unveiled Graviton3 at its late-2021 re:Invent conference, revealing that the 55-billion-transistor device includes 64 Arm-compatible CPU cores, runs at a 2.6GHz clock speed, can address DDR5 RAM with 300GB/sec max memory bandwidth, and employs Arm's 256-bit Scalable Vector Extensions.

The chips were offered as a tech preview to select customers. And on Monday, AWS made them available to all-comers in a single instance type named C7g.

EC2's C-series instances are billed as ideal for compute-intensive tasks. The series is now in its seventh generation – and the only seventh-gen instance uses Graviton3. Intel Xeons and AMD Epycs are currently confined to sixth-gen instance types.

Take that, you x86 dinosaurs.

The C7g instances offer eight sizes, with 1, 2, 4, 8, 16, 32, 48, and 64 vCPUs.

The Register has looked up hourly prices for the C7 instance, the C6 instance that uses the Graviton2, and the C-series instances running the third-gen Xeon and Epyc processors. We used the US West (Oregon) region as our guide, as it is one of just two regions currently offering the Graviton3. Our research considered instances running the same number of vCPUs offered by the C7 instance type.

As the table below shows, Graviton3 instances cost more than Graviton2-powered rent-a-servers, but less than their x86 competitors. AWS offers other ways to consume its infrastructure at lower prices.

vCPUs C7g Graviton3 C6i 3rd-gen Xeon C6a 3rd-gen Epyc C6g Graviton2
1 $0.036 N/A N/A $0.034
2 $0.073 $0.085 $0.077 $0.068
4 $0.145 $0.170 $0.153 $0.136
8 $0.290 $0.340 $0.306 $0.272
16 $0.580 $0.680 $0.612 $0.544
32 $1.160 $1.360 $1.224 $1.088
48 $1.740 $2.040 $1.836 $1.632
64 $2.320 $2.720 $2.448 $2.176

The AWS announcement of Graviton3 going into general production mentions that Snap Inc, operator of Snapchat and Bitmoji, plans to adopt the new silicon because it has proven faster than Graviton2. That is not great news for Google Cloud, as Snap signed a five-year, $2 billion deal with the ad giant in 2017. Five years later we discover that the social network operator has gone multi-cloud!

AWS's post announcing the C7 instances points out that migrating to Gravitons from x86 isn't hard for many workloads – especially if those workloads are AWS services like its own container and database services.

Which is utterly self-serving, but also an indicator that, as AWS brings more of its own services to its own silicon, trying Graviton silicon will become less onerous.

The timing of AWS's Graviton3 announcement appears not to have been coincidental, as it came a few hours before Nvidia revealed server designs based on its own high-end processors.

Perhaps AWS hoped to steal some of Nvidia's thunder.

Yet the combined effect of the two announcements may be more significant. Together they demonstrate that enterprise computing buyers can now consider multiple mature, well-backed technology choices beyond the x86 architecture. And there is more to come as Microsoft, Oracle, and others introduce their own Arm-powered servers, not to mention the imminent arrival of SmartNICs as accessible on-prem technology. ®

Narrower topics


Other stories you might like

  • This startup says it can glue all your networks together in the cloud
    Or some approximation of that

    Multi-cloud networking startup Alkira has decided it wants to be a network-as-a-service (NaaS) provider with the launch of its cloud area networking platform this week.

    The upstart, founded in 2018, claims this platform lets customers automatically stitch together multiple on-prem datacenters, branches, and cloud workloads at the press of a button.

    The subscription is the latest evolution of Alkira’s multi-cloud platform introduced back in 2020. The service integrates with all major public cloud providers – Amazon Web Services, Google Cloud, Microsoft Azure, and Oracle Cloud – and automates the provisioning and management of their network services.

    Continue reading
  • HPE thinks your next GreenLake deploy will be a private cloud
    Plus: IT giant expands relationship with Red Hat and SUSE, tackles hybrid data fabrics

    Extending a public-cloud-like experience to on-prem datacenters has long been a promise of HPE's GreenLake anything-as-a-service (XaaS) platform. At HPE Discover this week, the company made good on that promise with the launch of GreenLake for Private Cloud.

    The platform enables customers "to have a cloud in their premises wherever the data is, whether it's at the edge, it's at a colo datacenter, or is at any other location," Vishal Lall, SVP and GM for HPE GreenLake cloud services solutions, said during a press briefing ahead of Discovery.

    Most private clouds up to this point have been custom-built environments strapped together with some automation, he said. "It was somewhat of an improvement over the DIY infrastructure, but it really wasn't private cloud."

    Continue reading
  • Will Lenovo ever think beyond hardware?
    Then again, why develop your own software à la HPE GreenLake when you can use someone else's?

    Analysis Lenovo fancies its TruScale anything-as-a-service (XaaS) platform as a more flexible competitor to HPE GreenLake or Dell Apex. Unlike its rivals, Lenovo doesn't believe it needs to mimic all aspects of the cloud to be successful.

    While subscription services are nothing new for Lenovo, the company only recently consolidated its offerings into a unified XaaS service called TruScale.

    On the surface TruScale ticks most of the XaaS boxes — cloud-like consumption model, subscription pricing — and it works just like you'd expect. Sign up for a certain amount of compute capacity and a short time later a rack full of pre-plumbed compute, storage, and network boxes are delivered to your place of choosing, whether that's a private datacenter, colo, or edge location.

    Continue reading
  • Amazon can't channel the dead, but its deepfake voices take a close second
    Megacorp shows Alexa speaking like kid's deceased grandma

    In the latest episode of Black Mirror, a vast megacorp sells AI software that learns to mimic the voice of a deceased woman whose husband sits weeping over a smart speaker, listening to her dulcet tones.

    Only joking – it's Amazon, and this is real life. The experimental feature of the company's virtual assistant, Alexa, was announced at an Amazon conference in Las Vegas on Wednesday.

    Rohit Prasad, head scientist for Alexa AI, described the tech as a means to build trust between human and machine, enabling Alexa to "make the memories last" when "so many of us have lost someone we love" during the pandemic.

    Continue reading

Biting the hand that feeds IT © 1998–2022