Welcome to The Long View—where we peruse the news of the week and strip it to the essentials. Let’s work out what really matters.
This week, four things that caught my eye from Amazon Web Services’ re:Invent 2022 conference: SnapStart turbocharges Lambda, Graviton3E ARM HPC SoC, AWS continues hiring, and AWS’s origin story.
1. Lambda SnapStart Speeds FaaS
First up this week: Java Functions can be laughably slow, don’cha know? SnapStart aims to change that, by caching much of the repetitive startup required to bootstrap the runtime.
Steef-Jan Wiggers: AWS Lambda SnapStart Accelerates Java Functions
“Cached snapshots”
AWS announced an update to … Lambda, with the SnapStart feature, which reduces the cold start for Java Functions. [It] can, according to the company, sometimes take as long as ten seconds [to] bootstrap the runtime for the function.
…
With SnapStart enabled for a function, publishing a new version of the function will trigger an optimization process [creating] cached snapshots. … AWS states that Lambda SnapStart is ideal for synchronous APIs, interactive microservices, or data processing applications. … Currently, Lambda SnapStart is available in the US East (Ohio, N. Virginia), US West (Oregon), Asia Pacific (Singapore, Sydney, Tokyo), and Europe (Frankfurt, Ireland, Stockholm) regions.
Java schmava. What about other runtimes, Deepak Singh?
Java was a clear place to start … because the cold start problem with Java functions is significantly more pronounced than with other ones. But SnapStart is not … Java-specific.
…
Now that it’s out there, once we get a better understanding of how people are using it, we’ll find out which is the next runtime they want us to add support for and we’ll do that. We want to get to a world where you don’t have to think about cold starts.”
2. Graviton3E: ARM for HPC
AWS’s “Graviton” ARM instances take another step forward, with a couple of types tuned for high-performance computing. But where’s Graviton4?
Frederic Lardinois: AWS launches Graviton3E
“35% better performance”
At its traditional evening keynote at re:Invent, AWS [announced] a new version of its custom Arm-based Graviton chips. … These new chips will obviously power new AWS EC2 instance types. … All of these new instances will make use of AWS’s new Nitro 5 hardware hypervisor, which the company also announced. [It] promises significantly improved latency, up to 40% better performance per watt, and 60% higher PPS.
…
Graviton3E … was specifically designed for powering high-performance computing workloads. [It] promises significant performance improvements, including 35% better performance for workloads that heavily depend on vector instructions.
With a bit more detail, here’s Doug Black:
AWS said its Hpc7g instances … powered by new AWS Graviton3E chips, offer up to 2x better floating-point performance compared to … C6gn instances and up to 20% higher performance compared to … Hpc6a instances. The instances support CFD, weather simulations, genomics, molecular dynamics and other HPC workloads for clusters up to tens of thousands of cores. The instances have … 200 Gbps of Elastic Fabric Adapter (EFA) network bandwidth.
…
AWS said since the introduction … in 2013, the company has developed multiple AWS-designed silicon, including … three generations of Graviton chips. … AWS uses cloud-based electronic design automation as part of an agile development cycle for the design and verification of AWS silicon.
3. Hiring Freeze? What Hiring Freeze?
AWS says it won’t stop hiring—unlike the rest of the Bezos empire. And it most certainly won’t stop building new data centers.
Matt Day: Amazon’s Cloud Unit Plans to Add Staff
“We’re gonna keep building”
Matt Garman, a senior vice president who oversees Amazon Web Services’ sales and marketing teams, said he expected both his organization and the wider AWS business to add staff in 2023: … “Our business is still growing rapidly.”
…
A hiring freeze elsewhere in the company hasn’t derailed investment plans for its most profitable business. … “We’ll moderate our data center growth when the demand moderates,” Garman said. … “We have a lot of supply chain models that tell us to keep building data centers, so we’re gonna keep building them.”
Contextualize us, Tobias Mann:
Last week, Amazon announced it would spend $4.4 billion in India to add a second region. And as of this month, the company has more than 15 availability zones in development across Australia, Canada, Israel, New Zealand, and Thailand. The company also has five data centers, valued at $12 billion, under construction in Oregon.
…
AWS isn’t the only cloud provider continuing to expand its datacenter footprint in spite of worsening economic conditions. Microsoft this week said it would spend $1 billion on datacenters in North Carolina. … Google, meanwhile, has continued to expand to new regions in the Americas, Europe, the Middle East, and the Asia Pacific.
4. AWS’s Early History — ‘Written by the Victors’?
Here’s a snippet of Geoff Colvin’s fascinating tale: How Amazon grew an awkward side project into AWS
“That decision was the turning point”
For years AWS has brought in more profit than all other divisions of Amazon combined, usually by a wide margin. … AWS’s ascent is so unlikely that it demands an explanation. … How did this offshoot of an online retailer come to rule the lucrative cloud-computing industry? … The best-known origin story of AWS … won’t die, but it isn’t true.
…
By the early 2000s, Amazon … had built from scratch one of the world’s biggest websites, but adding new features had become frustratingly slow. … Teams were spending 70% of their time building the basic elements any project would require. … Every project team was performing the same drudgery. … Company leaders began to think, “Let’s build a shared layer of infrastructure services that all these teams can rely on.”
…
Amazon desperately needed to free its software developers. … Developers everywhere, not only its own, were starving for new tools that did just that. … But was that a business for Amazon? … That decision was the turning point. … On the day in March 2006 when AWS finally launched its inaugural service—S3 … 12,000 developers had signed up. … The revolution had begun.
To which, Microsoft’s Juan José Amor tweets:
I love to learn about successful business stories but this one in particular is touching me as I compete with AWS everyday. Kudos to every entrepreneur, to everyone making the right questions, to everyone willing to challenge status quo and for those who believe in themselves to shape the future.
The Moral of the Story:
The unexamined life is not worth living
—Socrates
You have been reading The Long View by Richi Jennings. You can contact him at @RiCHi or tlv@richi.uk.
Image: Masood Aslami (via Unsplash; leveled and cropped)
