Everyone needs a break from thinking about the incessant stream of new technology and Thanksgiving is a perfect time to reconnect with the natural world. The following are some highlights of my excursion into a beautiful wetlands nearby.
We’re all familiar with how Arm processors enabled the mobile computing revolution and are now disrupting cloud and desktop computing. Recently I had the opportunity to work with Bamboo Systems, a startup whose goal is providing the same Arm-powered benefits to enterprise data centers. I was a panelist in two webinars at GestaltIT where Bamboo described its company history and goals for a vibrant enterprise Arm ecosystem. In summarizing my views, I wrote,
Bamboo uses Arm efficiency to rethink enterprise servers…
Bamboo takes the same approach to servers by putting a high-end, 16-core A72 SoC on a high-density system board with up to 64 GB of memory that allows integrating four server nodes with a switch fabric into a narrow N-blade. The design includes a non-blocking switch driving an internal fabric with dual 10 GbE links to each node and dual 40 GbE uplinks to a ToR switch within each half-width blade that allows Bamboo to wedge two blades into a standard 1U chassis for an incredibly dense system.
As I concluded, Bamboo seeks to apply the benefits AWS is delivering via its Graviton Arm instances and Graviton-powered services to the enterprise.
There’s a reason AWS has ported its most-used services to Arm servers: they cost 40 percent less to operate for the same performance. The same cost-benefit analysis applies to enterprise data centers where Bamboo servers can provide the same or better for less up-front investment and far lower operating costs while reducing data center energy usage and associated carbon footprint.
For more details, read my column, along with impressions from the other analysts participating in the event, at GestaltIT and view the two webinars.
I also wrote a paper detailing the efficiency advantages of replacing conventional x86 servers with Arm-powered Bamboo systems using an model that addresses all factors of data center energy consumption. You can download the full paper here, but the following are some highlights.
Organizations are increasingly attuned to the fossil fuel usage required to power their internal opera- tions. Regardless of one’s view on anthropogenic climate change, no business likes to waste mon- ey by running inefficient equipment and since 60 percent of domestic electricity is generated using fossil fuels, the resulting CO2 emissions are signif- icant. With much of the debate centered on the shift to renewable energy sources, lost in the dis- cussion is the enormous energy and carbon sav- ings available from adopting more efficient tech- nologies. Perhaps nowhere in the business world are the efficiency opportunities greater than in the data center, which the IEA estimates (widely recognized as conservative) use about 1 percent of the world’s total electricity output.
Energy efficiency, often defined by computing performance-per-watt, is one area where Arm systems are indubitably superior to conventional servers. However, until recently, Arm processors could not deliver the baseline performance re- quired of data center workloads. Modern SoCs (systems on a chip) based on Armv8 and future Armv9 designs change the calculus by providing x86-class capabilities in a far more efficient pack- age. The advent of the Arm-based AWS Graviton processors, which by some estimates have already captured 20 percent of the instance share on AWS, demonstrates that Arm can run production workloads while delivering what AWS documents as “up to 40 percent better price performance over comparable current generation x86-based instances for a wide variety of workloads.”
We believe that there are many technical and economic reasons to adopt energy-efficient servers in cloud data centers, however, an unsung benefit is their potential for being both good for business and the environment. Indeed, we believe that Bamboo Systems is leading the way to the zero-carbon data center.
I greatly enjoyed an extensive discussion with my friend Vinnie Merchandani (@dealarchitect) on the many nuances underlying energy policy and why simplistic narratives around climate and the Earth’s ecosystems serve to distort the discussion and resulting policy proposals.
We discuss the slow progress at which renewables – solar and wind in particular – are scaling, Germany’s disappointing decade with them (see more in my note – The road to energy hell is being paved with good intentions ) and why the new US administration thinks it can do better. We also discuss why the French model with nuclear, the Iceland (and Kenya) model with geothermal, and the Norwegian hydroelectric model are not getting more attention.
We discuss emissions in Brazil, Russia, India, Indonesia China, S.Africa and other countries which continue with plenty of fossil fuels. Kurt has an interesting perspective on CO2 emissions. We discuss why the West appears to be shackling itself when their investment and progress away from fossils is being nullified by that in the developing world.
My goal is to take the discussion back to economic and scientific realities (and unknowns) and out of the political cesspool and cartoon characterizations in which talking points and pseudo-religious mantras replace reasoned debate.
AWS has grown into an IT version of its parent Amazon: a sprawling online marketplace where buyers can find anything they need. What is Amazon? A grocery store? Drugstore? Hardware and home center? Electronics mart? Book, music and movie store? Indeed, it’s all of these rolled into an app delivering a convenient, frictionless shopping experience. AWS has evolved into the same thing for IT services: a place with something for every need. Thus, predicting what AWS might introduce at re:Invent has become less interesting as it’s grown more predictable by annually providing more of the same.
- More instance types further tailored to different applications with a recent focus on AI development, model training and inference.
- Additional management features to automate cloud operations, security configuration and resource optimizations.
- New database configurations to accommodate those few enterprise apps not already supported.
- Better price-performance as AWS exploits improved processor, storage and network technology and, recently, by employing custom chip designs using alternative processor (Arm) and machine learning architectures.
Although AWS has become the all-in-one superstore for cloud infrastructure, it lags behind its largest competitors, Microsoft and Google, in products designed for the new, post-Covid hybrid workforce dependent on cloud productivity, collaboration, messaging and video services. While AWS has some SaaS products, what it calls Business Applications, services like WorkDocs, WorkMail and WorkSpaces (VDI) pale in comparison to category leaders like Office 365 and Google Workspace. Furthermore, these Business Applications don’t adequately address capabilities like group collaboration (Slack) and real-time communications (Zoom). A key selling point of Microsoft and Google when negotiating enterprise cloud deals is their ability to provide both infrastructure and workforce applications. Expect AWS to start closing this feature gap at re:Invent via a combination of new products and close partnerships with one or more category leaders like Box, Dropbox, Cisco (Webex) and Zoho.