Intel x86 vs. Arm: The Most Important Selection Criterion – New Futurum Research Report

The News: Businesses are prioritizing a successful cloud journey as not only essential to their long-term competitiveness, but also as a key driver for meeting organization-wide digital transformation goals. In order to streamline the journey to and across clouds, we believe organizations must attain a fundamental understanding of microprocessor architectures that underpin cloud ecosystems. That’s why we recently evaluated Intel’s x86 architecture in this white paper, where we explore what we believe to be the most important selection criteria for any microprocessor architecture.

While we are seeing and hearing an increased volume of interest and consideration for Arm based servers, we believe it is still very early days for the platform, and while we do expect Arm based offerings to improve over time, this will be an evolutionary process, which is why we see x86 as the leading platform for server technology for the foreseeable future. By understanding the built-in capabilities of the x86 architecture, businesses can gain the full benefits of flexibility, predictable performance, efficiency, and security in driving their cloud journeys.

Specifically, the x86 architecture enables organizations to swiftly adapt to changing business requirements including the ability to redirect workloads based on business factors such as cost, reliability, and security. Intel x86 assures predictable performance through providing customers choice of cloud infrastructure based on Quality of Service and Service Levels, eliminating highly variable results.

The Intel x86 architecture increases visibility into costs across data center and clouds, more efficient use of infrastructure, as well as grow and shrink cloud resources based on usage. The x86 architecture ensures the automation of security profiles, data sovereignty, as well as unified administration and control for regulatory compliance. In sum, through the Intel x86 architecture businesses can build the cloud optimized architectures best suited for their requirements, whereas Arm architecture does not yet provide the same assurances.

Intel x86 vs. Arm: The Most Important Selection Criterion – New Futurum Research Report

Analyst Take: In our newest research report, “Intel x86 vs. Arm: The Most Important Selection Criterion” — done in partnership with Intel — we examine why it’s important that organizations better understand the underlying microprocessor architecture that provides the foundation for the global cloud ecosystem. In this regard, Intel has a long history in advancing microarchitecture capabilities, including a track record of more than 40 years of proven refinement and the massive installed base of applications tuned for the Xeon microarchitecture ecosystem. Intel has driven generations of Xeon microarchitecture optimization through input and lessons learned from its installed base.

From our perspective, Intel’s Instruction Set Architecture (ISA) ensures the basic instruction set in any architecture is the same as evidenced by decades of customer and application driven ISA extensions. Emerging workloads are optimized with new ISA inputs providing significant performance upside from added ISA when needed.

We believe the consistent refinement and advancement of the x86 microprocessor architecture has assured the streamlining of core workload distribution and management across all cloud environments, avoiding the complexities and costs of running multiple workloads in the cloud. Of critical importance, we point out that virtually all robust, proven intellectual property (IP) is running on the Intel x86 architecture.

As such, moving or refactoring an x86 architecture to an alternative Arm architecture can potentially be a costly, fragmented undertaking. The adoption of Arm-based processors, such as AWS Graviton, is not a five easy steps program. Likewise, when we explore the virtualization of x86 alternatives, such as the AMD portfolio, we see similar risks posed, with the potential of disrupting business continuity while also requiring a significant escalation of resources and time to administer.

For instance, writing code that resolves compatibility issues with a new architecture is inherently time consuming. Also, specific hardware, such as Arm-based Graviton chips, and supporting toolkits, including compilers and libraries, may not be readily available to developers, creating delays in the overall porting process. Moreover, we see the ability to predict the performance of software before conducting tests on production chips is difficult and adds uncertainty.

Key Takeaways on The Future of Cloud is the Network

In sum, pursuing microprocessor architectural diversity increases costs, headcount, and overall support structure. In our opinion, the stakes here are too high for businesses to risk the disruption of their workload distribution across any cloud environment — including private, public, hybrid, and multiple. From our perspective, when it comes to sustaining and advancing the microprocessor architecture of the cloud ecosystem indulging engineering hobbies is an imprudent business choice. Download the full report here: Intel x86 vs. Arm: The Most Important Selection Criterion.

Disclosure: Futurum Research is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.

Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of Futurum Research as a whole.

Other insights from Futurum Research:

Intel x86 vs. Arm: The Most Important Selection Criterion

Intel Vision 2022: Intel Augments its Silicon, Software, and Services Portfolio Vision

Intel Q1 2022 Earnings Beats on Top and Bottom Line, But Guidance Continues to Be a Sticking Point

Image Credit: Intel

The original version of this article was first published on Futurum Research.