The Arm architecture dates back to Acorn Computers’ BBC Micro in 1981. The BBC Micro was a conventional machine, yet it ran nearly twice as fast as its contemporaries, such as the Apple II. About the same time, research on a microprocessor design based on the Reduced Instruction Set Computer (RISC) demonstrated that simple chip designs could easily outperform the current high-end 32-bit offerings.
This pivotal moment in computer history inspired Acorn to release its own CPU design, the Acorn RISC Machine, in 1983. These computers would eventually become known as Advanced RISC Machines, popularized under the acronym ARM.
Arm architecture is still a popular choice used in a range of computing applications. In this article, you will learn about the features of Arm architecture, compare it to the x86 architecture, and explore the benefits and drawbacks of using Arm in the cloud.
What is Arm architecture?
Arm is one of the most ubiquitous architecture types used in computing. Although it is widely used, it can be surprisingly difficult to understand what it is and why it is so popular.
Arm architecture belongs to the RISC design family. RISC microprocessor architecture design uses a highly optimized set of instructions that enable a small processor to efficiently handle complex tasks.
Arm has become the cornerstone of mobile devices, the world’s largest computer ecosystem, and many experts consider Arm the future of cloud computing. Reasons for its growing popularity include its low power consumption, flexible licensing, and low costs.
Low power consumption
Arm provides excellent performance per watt when compared with other microprocessor architectures. Its advanced version of the RISC design offers outstanding performance while maintaining superior energy efficiency. This efficiency is crucial for battery-operated devices like smartphones, tablets, and laptops, allowing them to last longer while maximizing performance speeds.
Fair and open licensing
Arm architecture’s license provides significant user control. The license is for intellectual property instead of specific code or hardware. As a result, end users enjoy generous flexibility when using the Arm architecture.
For instance, Apple uses Arm architecture to develop processors optimized for its product-specific functions. Apple recently revolutionized one of its Arm-reliant processor groups with the new M1 CPUs.
Low cost of entry
Arm licensing is reasonably cheap compared to what other companies charge for their processors and architecture. Arm does not manufacture physical chips and thus has lower overhead.
Any company can add its own system-on-chip (SoC) technologies to Arm architecture. This process lets users customize the Arm architecture to their proprietary chips, optimizing them for specific uses. As Arm grows, the reduced manufacturing costs inherent in the ecosystem allow for continually increasing cost efficiency.
The low power consumption, fair and open licensing, and low cost of entry combine to create an ideal platform for mobile and cloud computing. However, Arm is not without its competitors, especially in the cloud computing market. Arm’s main competition in this space is Intel’s x86 architecture.
Arm versus x86
Intel first introduced x86 in 1978 and debuted the now-standard 64-bit version in 2003. The x86 architecture design dominates devices such as personal computers and game consoles.
For a long time, x86 also commanded the high-end compute-intensive workstations and cloud computing markets. Circumstances are beginning to shift, though, with spaces like Amazon Web Services (AWS) offering many Arm-based cloud computing options. Even today’s fastest supercomputer is Arm-based.
Market tides are turning, and Arm appears to be breaking ahead of the pack.
Differences between Arm and x86
The primary difference between Arm and x86 processors is the way that they execute their instructions. x86 systems are based on complex instruction set computer (CISC) designs, while Arm is based on RISC designs. In short, x86 processors can execute multi-step operations on each instruction, while Arm processors use a limited, but highly optimized, set of instructions.
This execution difference is not the only reason that Arm is beginning to outperform x86 in important markets. Arguably, that change is due equally to Arm’s attractive price-to-performance ratio.
For example, consider the AWS Graviton processor, Amazon’s first AWS Arm offering. Arm can save AWS users as much as 40 percent in costs for performance equal to that of Amazon’s previous x86 cloud processors.
Arm’s flexible licensing process comes into play as a critical difference between Arm and x86. Intel’s x86 processors are still a proprietary chip, meaning that Intel is the sole creator of the physical hardware and sole decision-maker regarding the chip’s SoC. This is a limiting ecosystem when compared to the Arm licensing scheme. Companies may not be able to optimize the abilities of their processors and specific use cases when relying on x86.
Benefits and drawbacks of Arm in the cloud
Besides Amazon, many other organizations are choosing Arm architecture and not x86 for their cloud systems. Arm may be the right choice for you too, but there may be some disadvantages of using Arm in the cloud. It may be helpful to review Arm’s positive and negative attributes.
Superior price for performance
Price for performance is a deciding factor for companies that choose Arm, and for good reason. Arm is just as efficient as x86 for cloud computing while still offering attractive prices.
Organizations can save money while processing a broad spectrum of workloads, truly putting Arm’s performance into competition with x86 architectures. In this area, its rivals simply cannot surpass Arm’s price-to-performance ratio.
Arm’s licensing has enabled companies like Amazon to integrate the architecture into their cloud-focused chips, including the Graviton2 CPU. Using this integration, Amazon can build on the Nitro Security Chip, offering unparalleled security in the cloud computing market.
The Nitro chip offers 256-bit DRAM encryption and 50 percent faster per-core encryption performance. The proprietary nature of x86 chips is too restrictive to have enabled Amazon to benefit from this innovation.
One drawback of Arm-based cloud computing is the reduced compatibility of certain apps or dependencies. The historically dominant presence of x86 hardware in PC and cloud spaces means that developers have built most apps and containers to run on x86 machines. You may need to recompile and reconfigure your team’s apps and dependencies for Arm-based systems.
x86 has always been the foundation of cloud computing. Arm is still a relatively new offering for most cloud suppliers. So, while Arm instances are available for Amazon Elastic Compute Cloud (EC2), Oracle Cloud, Alibaba Cloud, and a few other spaces, they are not yet present on Google Cloud Platform (GCP) or Azure. Availability may limit your Arm adoption if you rely on Google Cloud and its ecosystem or if you need to switch platforms to continue your product development.
Is Arm the future of cloud computing as we know it? The answer is that it depends.
Arm is part of the future. Whether you can use Arm exclusively depends on your specific needs and workloads. Although Arm is a leader in price-to-performance ratio and security, it may cause issues when integrating with certain applications or platforms.
That said, in many cases, you can move your workloads from x86 to Arm with little to no modification. Overall, Arm is an exceptional platform that is expanding the potential of cloud computing, just as it did with mobile devices decades ago.
If you decide to use an Arm-based system in your next cloud project, make sure that you have access to effective continuous integration and continuous deployment (CI/CD) tools. Good CI/CD can ease your transition to Arm by simplifying the building and deployment necessary for Arm-powered virtual machines (VMs). To begin your exploration, check out the offerings of CircleCI’s robust Arm support. If you do not have a CircleCI account, sign up for a free one here.