Skip navigation


By Stephen Turnbull, Director of Marketing, Vertical Segments


AMD has experienced success in numerous embedded computing markets, perhaps none more so than with “thin clients.” They are thin because they have little or no local storage and often serve as intelligent front-ends for server or cloud-based applications. This success reflects our position of offering high-performance thin clients for demanding applications. In particular, AMD thin clients have a strong value proposition for visual computing, which is any business-class embedded system that mixes computation with single- or multi-display capabilities at a compelling performance per dollar per watt.


AMD is a significant player in the thin client market and is well positioned for continued growth given that AMD has achieved the leading market share for thin clients, based on internal reports, and continues to sign deals with major thin client manufacturers.


Success comes when we leverage innovative technologies across multiple market spaces. Think of the AMD Accelerated Processing Unit (APU). We’ve seen success in the traditional client space with performance desktop and notebook, and we’re seeing increasing interest in the thin client space.  So, what type of thin client applications would need an x86 AMD APU? A couple of key markets stand out. Local, state and federal governments are a big user. Why? A key advantage for AMD thin clients based on the Embedded G-Series APU is built-in security. Each APU includes an AMD Secure Processor based on ARM® TrustZone® architecture with an independent ARM Cortex®-A5 processor, eFused keys and a crypto co-processor. This allows functionality such as Trusted Platform Module 2.0 that stores RSA encryption keys specific to the host system for hardware authentication ensuring that deploying thin clients does not mean adding points of weak security to a user network.


Banking is another area of strength for thin clients with an AMD APU. For bank tellers and even professional bankers, there’s no need for local storage at the client. What is needed is fast processing and strong visuals to graphically depict account status and financial products. The strong graphics performance of the native APU or with an AMD Radeon™ discrete embedded GPU can drive dual 1080P screens with the pending support for dual 4K including H.265 video. Other display technologies from AMD include AMD Eyefinity technology, which allows a single image to be spread over multiple screens, and “dual graphics,” which allows an APU and discrete GPU (dGPU) to work together to provide higher graphics performance. This is a significant benefit in the financial services market, including stock brokers and traders who often work with multiple, high resolution displays.


Recent design wins with FujitsuHP, and Samsung validate that thin clients using an AMD APU provide more-than-ample horsepower for data movement, encryption/decryption of central server data, and even local on-the-fly video encode/decode for video conferencing or multimedia streaming. Besides financial services and government, other users include call centers, kiosks, and hospitals, “smart” monitors and TVs, military command posts and other multi-user, virtualized installations.


AMD provides impressive performance per dollar per watt, strong security, sophisticated power management, and superior graphics performance. The product lineup includes an unparalleled range of pin- and software-compatible offerings, helping to address multiple model needs for our clients. AMD has what customers want from their thin clients.


Stephen Turnbull is Director of Marketing, Vertical Segments, for AMD. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

Security That Locks You Down, Not In


Whoever sang that the holiday season is “the most wonderful time of the year” probably wasn’t an IT pro facing a slew of malware, cyber espionage, and other targeted attacks. As millions of consumers flock to grocery stores, boutiques, and websites to complete their holiday shopping – and millions of businesses gear up to take, fulfill, and ship orders – attackers are gearing up to infiltrate personal and business-critical transactional data. With cybercrime becoming increasingly sophisticated, is it naive to rely on business as usual without, inevitably, receiving a Pandora’s Box of malicious agents?


At AMD, we understand that cybercrime is a persistent, 24/7/365 threat. We understand that anti-virus measures alone aren’t enough, that both businesses and consumers need equally persistent defenses against this threat. And we understand that constantly improving data security starts with securing the data center, integrating state-of-the-art software with innovative hardware that protects from the silicon up. If you haven’t heard, we’ve been busy driving this forward in two big ways: the advent of AMD Secure Technology and alignment with ARM® TrustZone®.


AMD Secure Technology offers a built-in security system that puts protection right on your processor. For companies with secured infrastructure, this technology can mean heightened security for cloud-based servers that helps protect corporate data and manage remote workers’ machines.


ARM TrustZone technology is a system-wide approach to security, isolating and protecting sensitive applications and interfaces to protect services and devices from scalable attacks. Building on open-standards-based architecture and expanding our collaboration with an extensive network of platform providers, we’re working to provide the greatest peace of mind on every AMD product.


Case in point: the AMD Opteron™ A1100 processor (formerly codenamed “Seattle”), AMD’s first 64-bit ARM-based processor. Each AMD Opteron A1100 processor has a dedicated 32-bit microcontroller, isolated on-chip ROM and SRAM, and is 100% ARM TrustZone compliant. When you prioritize security at the hardware level, this instills confidence about capabilities, from storage to web serving to software development. Judging by the expanding ARM64 development on AMD by innovators such as SoftIron, Inc. and CircuitCo, we’re contributing to a healthy, growing ARM ecosystem.


The reality is that cyber-attacks won’t stop, but neither will our determination to equip you with server technology that’s deeply secure by design, hardware that works hand-in-hand with software to help protect your business, your customers, and your future.


  • The AMD Server Team

norman_fraser_softiron.jpgUntil recently, ARM has been notable by its absence from the server market. But the last couple of years have seen the arrival of 64-bit ARM-based server processors such as Applied Micro’s 8-core X-Gene, using the 8-core AMD Opteron A1100 processor, and Cavium’s 48-core Thunder-X, each of which supports vital enterprise features such as ECC memory and virtualization.


Some may grumble that first-generation ARM server processors are not the cheapest or the most powerful or (no kidding!) made by Intel, but it is nothing short of remarkable that the first generation of a brand new class of server silicon is already finding niches that matter to users where it out-performs Nth-generation x86 silicon.


Of course, x86 will continue to be a major player in the server arena. But though it possesses impressive product maturity and market share, it lacks the structural advantage of ARM’s licensing business model, a model that offers more diversity and choice for better-fit solutions. In addition to Applied Micro, AMD, and Cavium, other heavyweight players look set to bring their own ARM-based 64-bit server SoC ranges to market, and we see ARM as an opportunity to grab a share of the sizeable server market.


To date, ARM has licensed its designs to more than 1,200 partners,1 with awesome cumulative market reach. This private army of innovators has powered ARM’s seemingly inexorable advance into new markets and supercharged its market share growth. The number of chips containing ARM processors shipping per year grew from a few hundred million in 2000 to more than 12 billion in 2014. 1 With that kind of track record, I believe it would take a brave observer to bet against ARM succeeding in the server arena.


The strength, number, and reach of ARM’s licensees are key to why we at SoftIron believe ARM has a better chance of success than PowerPC. The ARM community has an enviable track record of grabbing market share once product meets the functional criteria required in each market niche to be colonized. Data centers prove the point. ARM processors are already pervasive in the data center, in the shape of storage disc controllers, environment monitors, and a host of other instrumented functions, where attaining the necessary functional criteria led inexorably to market share growth. We believe the same process is now beginning with respect to the data center’s headline act: the server CPU.


At SoftIron, we’re proud to be amongst the first to launch 64-bit ARM-based products into the market and to be partnering with silicon innovators like AMD. Our experience of selling into the emerging market gives us an almost unique perspective on what is actually driving ARM server interest and sales.


Sure, people are interested in performance per watt, packing density, thermal footprint, and total cost of ownership. These matter, and the arrival of ARM in the data center brings new options, especially as adopters look to improve operational efficiency through workload optimization. But what ARM server adopters want most of all is choice. Choice to build redundancy across more than one server architecture. Choice to be a little less dependent on the dominant silicon vendor. Choice to explore for themselves where x86 makes sense and where ARM offers better performance against KPIs.


Norman Fraser PhD is Co-founder and CEO of SoftIron, a leader in ARM-based enterprise computing products with bases in England and Silicon Valley.


1 ARM Holdings, Q3 Roadshow Slides,