VDC Research: Embedded Microprocessor, Board & Systems Market Blog

Intel stacks more chips onto its existing M2M wager

The wagers being played in the high-stakes M2M game are definitely getting higher. As we noted last week, Eurotech, who has long being seen as a leader in edge gateways and M2M solutions, divested its Parvus asset in order to place increased focus on its core markets.

Yesterday, Intel announced a commitment to deliver intelligent gateway solutions in early 2014, offering a combined platform of Intel, Wind River and McAfee products.   These new, pre-validated products will allow embedded computer suppliers and their customers to more easily support M2M deployments targeted at legacy equipment as well as new products. The Intel M2M gateway solutions will be scalable from small deployments of a few supported things using Intel’s new Quark or Atom processors to large-scale deployments driven by higher powered Xeon products.

Why are companies such as Intel and Eurotech so focused on gateways? VDC’s previous and continuing coverage of scalable edge computing products has the answer. Unlike many markets, there is and will continue to be solid growth in all regions as the Internet of Things (IoT) and M2M become increasingly vital for saving operational costs and providing new revenue streams.

In 2014, VDC will again be reporting on intelligent edge node gateways. We look forward to your company’s participation as we shape the report scope and conduct the required research. For more information contact us at [email protected]

Will the US Government Shutdown Impact the Embedded Computing Market?

The answer is yes and depending on the length of the disruption, the impact will be significant. Decisions on new US aerospace and defense projects will be on hold but the effects of these delays will be minimal as these decisions only impact embedded computer supplier revenues 3 or more years in the future.

What will be significant are the numerous small impacts over multiple markets which will tend to move embedded computer supplier revenues that would have been recorded in 4Q/2013 to 1Q/2014. Delays in inspecting and approving products at or for OEMs in medical, aerospace, and defense markets will ripple back through the supply chain and therefore will impact embedded board and systems suppliers first and then processor and memory suppliers. If the inspection delays hit the US ports, the consumer electronics market will be the next to suffer and that will represent a much bigger impact that will eventually hit industrial markets.

The shutdown of the National Institute of Standards and Technology (NIST) will also be problematic. This will be particularly true if US based OEMs and semiconductor manufacturing or test facilities already or soon need their master standards to be calibrated by NIST. The effect will be minimal at first until the given facility has a critical mass of instruments and test equipment on the manufacturing line that have to be calibrated.

If the shutdown quagmire extends into the debt ceiling deadline on October 17, the impact will be far greater as revenues will be lost as opposed to delayed. In defense markets, it has been stated that ‘flat is the new normal’ but, if the US continues on the path it is following, flat may be the fortunate exception and that logic may well apply to many markets not just defense.

Curtiss-Wright is Betting Big on SWAP

In today's earlier blog covering Eurotech’s big bet on M2M, we mainly discussed the Eurotech side of the deal. In this follow-up we will provide VDC's View of the Curtiss-Wright side as we believe they have a shrewd strategy in play. In Parvus, Curtiss-Wright receives a complementary set of mobile computing and networking products that will allow diversification beyond their core Mil/Aero markets into markets such as industrial where ruggedized products with superior Size, Weight, and Power (SWAP) attributes will be increasingly valued. With ~$20M or more in 2012 revenues and an EBITDA of ~25% Parvus is definitely not a white elephant.

In summary, we see the basic business principles behind the Eurotech / Curtiss-Wright deal as being similar to those behind the 2012 blockbuster deal between the Boston Red Sox and Los Angeles Dodgers. The Dodgers acquiring  valuable players complementary to their target market and the Red Sox gaining liquidity to pursue a different team building strategy. Given that 2013 finds both the Red Sox and Dodgers well positioned to succeed in the playoffs, one can hope that Eurotech and Curtiss-Wright will also succeed in their pursuits in 2013 and beyond.

Eurotech is Betting "All-In" on M2M and the Internet of Things

Today’s announcement of the Eurotech sale of its Parvus division to Curtiss-Wright represents an all-in bet by Eurotech on the growing Internet of Things (IoT) and M2M market in general.  The approximate $38M in capital will allow Eurotech to further develop and support its portfolio of the High Performance Embedded Computing (HPEC), and “green” highly-efficient super computing products required at the edge of device networks as well as the cloud resources that support them.

In the next 3-years we can expect that strategies and actions such as those undertaken by Eurotech will be the rule rather than the exception. One limiting factor for rapid M2M adoption is that the billions of things that will be connected communicate over a myriad number of protocols, frequencies, and methods with diverse data structures.  At the same time, there are often strict security requirements to be followed. In these cases, gateway devices such as Eurotech’s Reliagate platform perform the necessary conversion, aggregation, and security functions.  Eurotech is not alone with their increased focus on M2M gateway platforms as they collaborate with IBM in support of the MQTT protocol used in gateway middleware deployments. Competition is increasing for edge computing software and hardware however. At Oracle’s Open World event last week, Freescale and Oracle announced a One Box platform collaborative product that includes Freescale processors and Java SE and Java Embedded Suite.  The OneBox platform is directly target at M2M gateway applications in consumer/home, medical, mobile, and industrial markets.

The Great Divide: Growing Armies for High-end Server Applications

Several recent reports stemming from the keynote at least week’s JavaOne conference indicate that IBM is working towards enabling GPU acceleration with Java, one of the most popular programming languages used in software development. This shortly follows the establishment of the OpenPOWER Consortium, an open development alliance based on IBM’s Power architecture, in which an announcement was made that IBM and Nvidia will work together to integrate the Power and CUDA GPU ecosystems. At this point, its hard not to think about the competitive strategies unfolding of other industry leaders such as AMD, ARM, and Intel to capture the burgeoning server markets for high-performance computing (HPC) and big data applications. The end result will determine more than who carves out the most revenues, it will also decide the fate of several processor architectures and their extent in new enterprise applications.

First, from a software standpoint, IBM and Nvidia’s recent announcements will surely attract and retain a lot of developers. Not only is the collaboration enabling two pools of (typically deeply-invested) software developers to pursue more application opportunities, it will also add support for a common high-level language to which OEMs can easily acquire or build out expertise, Java. Nvidia is no newcomer to GPU acceleration in demanding big data applications, and teaming up with IBM to offer a more-holistic solution featuring more than its own proprietary CUDA language and GPU technology will improve integration with prospective businesses’ requirements and objectives.

On the other end, we have the splintering relationship between AMD and Intel. Traditionally an x86-only supplier, AMD will instead be also rolling out the first 64-bit ARM processors for datacenters in 2014, and will supply “processor-agnostic” SeaMicro servers and data center technology. Despite the sudden, uncomfortable ripples by AMD (whom will continue to supply x86 server chips as well), Intel’s x86 ecosystem is still widely used in datacenters around the world. Furthermore, the company has its own line of computing products to compete with GPU acceleration – the Xeon Phi coprocessors, launched in mid-2012. Intel’s processors and coprocessors use common languages, models, and tools – maximizing developer support and preserving software expertise across the Xeon product line.

IBM and Nvidia’s growing relationship is hugely important for each company. Neither on its own would likely make a significant impact in the markets for large and hyperscale computing with respect to Intel and ARM’s growing influence. However, bridging popular CPU and GPU architectures will bode well for CUDA and Power as OEMs increasingly employ heterogeneous architectures to realize the flexibility in simultaneous processing of serial and parallel workloads. 

Keep Calm and Carry On, Discrete DSPs

Processor technologies are undeniably moving towards consolidated, integrated architectures to take advantage of multiple processor types in tandem while saving on size and power consumption. While such heterogeneous ICs are changing the way chipsets are developed, licensed (core IP), and manufactured, not all applications need or necessarily want them. High-performance remains the chief metric by which end users select embedded processors, particularly for those using digital signal processors (DSPs). As a result, intensifying application requirements will preserve the market for discrete DSPs in a world increasingly prevalent with holistic system on chips (SoCs).

Signal processing requirements are escalating in a variety of industries. The continued expansion of 4G wireless technology throughout Europe, as well as increased 3G adoption in emerging Asia-pacific countries such as India, will fuel growth for high-performance DSPs in base stations and other access layer equipment. The greater demand to provide sufficient cellular quality of service and coverage is amplified by the tremendous growth in connected devices produced by the Internet of Things.

Though upgrading and deploying new backhaul networks will be the leading application for modern and next-gen DSPs, high-performance signal processing will still be crucial in other industries, such as consumer electronics and industrial automation. For example, the new Apple iPhone 5s includes a secondary coprocessor, the M7, embedded with DSP technology specifically designed to easily manage and migrate sensory workloads from the main processor – simultaneously increasing device performance and saving energy. Similarly in the industrial automation space, the growth in peripherals on the shop floor such as cameras and sensors for machine vision applications will necessitate the parallel processing performance of high-end DSPs.

Though several discrete processor markets continue to combat SoCs for design wins (including DSPs), the proliferation of heterogeneous processors has allowed traditional DSP suppliers to form new revenue streams through IP licensing, much like what is already done for CPU and GPU cores. Furthermore, new and growing application requirements will still require pure, optimized DSPs for the foreseeable future. For these reasons, we believe discrete DSP manufacturers can keep calm and carry on.

Calling all Embedded Board & Module Suppliers – Stand up and be Heard!

The winds of change are blowing through the embedded computer board and module market. As 2014 approaches, recent market consolidations and reorganizations will solidify. 2013 will go down as one of the most important years for suppliers to make their voices heard by VDC. Why is this so important? In 2014 and beyond there will be significant shifts in demand for embedded board and module products as seen by these results from VDC's 2013 survey.

Many OEM engineers will be designing their new products and, for the first time they will be using embedded modules as opposed to other product alternatives. At the same time, existing module users will be looking to migrate to the latest x86, ARM and/or FPGA processors and module types. These OEM engineers will value an authoritative 3^rd party source to better allow them to determine which product types are strongly supported and, which product types not gain the necessary market traction.  On a similar note, these OEM engineers and other entities with interest in the embedded market are looking to VDC to determine which suppliers are becoming the most resilient, innovative, and successful in the markets they serve. VDC provides this industry-wide guidance by taking a 360 degree view of the market but, even so, the supplier inputs we receive are, among the most valuable.

At present, VDC has been in contact with over 90 suppliers we believe are significant to the market and after many initial discussions have provided our top-level estimates for their company. We are following these e-mails and calls with our 2013 survey that has detailed estimates. If you have NOT received these e-mails, calls and survey and, you believe your company should be represented, please contact us immediately. If you have received this survey it MUST be received by Friday October 4 to ensure any corrections or fresh guidance are included in the VDC models that drive our reports.

 

Will the Hynix Wuxi Fire Impact the Embedded Computer Market?

We expect the September 4^th  fire at a Hynix semiconductor fab producing DRAMs to adversely impact suppliers of embedded products. The perception of a possible supply disruption for DRAMs has already affected pricing. Although Hynix is downplaying the event, the likely affect will be somewhere between a temporary shortage and a larger industry impact that some investors are predicting. Suffice it to say that the potential mix between a cleanroom and “thick black smoke” would be disrupting, even if the equipment was not obviously damaged. It is possible that materials and surfaces in some parts of the facility were contaminated. If so, the effects of semiconductor or electronic circuit board contamination may take months or even years to appear and this would have two possible impacts:

• Some of the Hynix production equipment may become less reliable, which can impact supply.
• Some of the components Hynix produces at that facility may have higher failure rates, particularly over the longer time frames in which embedded computers are deployed.

Whether either of these two possibilities will actually happen may be immaterial as embedded board and system suppliers are a cautious bunch who may choose to lock in supplies of DRAM products from alternative suppliers.

3D Printing Will Impact Embedded Computing? VDC’s View of the Future

Technologies are evolving quickly in 3D printing and new nano-materials. Will these new technologies impact the market for embedded computing products? We believe the answer is, yes, but, not immediately. The 3D printing process is too slow for the production volumes needed for many if not most products that contain embedded computing. The first embedded market impact of new 3D and nano technologies will be on the engineering processes that create products not the products themselves.

Is the concept of printing electronic circuits new? Not in the slightest. When I first began my engineering career in the late 70s, the company I worked for produced laser trim systems. These systems were used by customers to measure and perfectly adjust the resistors on ceramic modules that had been printed using a silk screen printing process.

Exhibit 1: Screen Printed Thick Film Ceramic Module

After process steps for printing the circuit traces and resistive components, the laser system would then be used to make continual measurements as laser pulses cut into the resistive material. This would increase the resistance until it met the target value. Prior to laser technology, this same task would have used a sand blast method. In either case, the modules would be completed by adding semiconductor dies and encapsulating the module.

Printed circuit boards (PCBs) are not really printed as much as they are plated and etched. Because of the density and complex circuits multiple layers are usually required. PCBs like the 70’s era ceramic modules require significant manpower and expense to design and layout the circuits and then create the necessary screens and photo masks needed to produce the product. Even with all the automated tools that engineers can utilize, this PCB development process can take a significant amount of time. If the first articles produced are found to have problems, a re-spin adds more time-to-market delay. It is worth noting that the equipment needed to create PCBs and embedded modules are expensive and therefore most companies rely on contract manufacturing specialists to produce them.

The same pictured module could be produced using a 3D printer either on top of ceramic or, perhaps, using another material as a substrate. The base might even be flexible, allowing the finished product to be more adaptable. By precisely controlling the amount of resistive material, a trimming process could possibly be eliminated. Even capacitors could, (in theory), be printed.

Most importantly, a small engineering team could be testing a prototype of a new embedded computer product within hours of developing it. Production of the verified design would still be completed by the aforementioned contract manufacturers, but the time-to-market would be more compressed and there would far less uncertainty about having to make engineering corrections. At a minimum, the ability to rapidly produce prototypes and proof-of-concept products can level the playing field for smaller less capitalized embedded computer suppliers.

In the distant future, as 3D printing capabilities and speed increase along with the development of nano materials that can be used as “ink”, embedded hardware may take on an entire new meaning. Consider an automotive part like a bumper that was manufactured with all of the electronics needed for a collision avoidance system completely embedded in the bumper’s material. As the bumper is printed, the circuit traces and components could be printed or picked and placed. Is this printing/embedded process feasible today? No, but like previous disruptive technologies its time will come.

What should embedded suppliers do now? As 3D and material technologies start to mature and stabilize, suppliers should begin to use 3D printing to produce prototypes of embedded sub-modules. This exercise should be conducted in parallel with traditional processes to test if and when time to market compression exists. When the new technologies are proven, suppliers can confidently develop new products because they will know that the engineering prototypes they produce will be reliable proxies for their completed products.

OEMs Shift Their Use of Professional Services to a Gain Competitive Edge

In 2013, VDC is seeing dynamic shifts in strategy for many OEMs as they respond to market pressures and position their organizations for future opportunities. As the economy improves, the calculus behind the professional services decision may change but most OEMs still prefer to outsource non-core functions. Although the OEM may be losing some product margin to pay for the services consumed, it is a small price to pay compared to the expense and risk of bringing processes back in house. Even so, we see that 10% of OEMs will be considering re-joining the 20% of their peers that do not utilize any professional services (Exhibit 1).

Exhibit 1: 2013 OEM cited current and future use of professional services

In these cases where the reduction in professional service use is being considered by the OEMs, there could be multiple forces and strategies at work. Sometimes it can be as simple as the OEM’s belief that if they want to do something right, they have to do it themselves. It also can be the result that recapturing the lost margins is worth the organizational costs and risks given the improving unit volumes. Although improving margins through insourcing may be part of the strategy, the largest gains are obtained when an OEM can better differentiate their product and therefore charge a premium price for it. Therefore, it is likely that OEMs with proprietary processes and technologies that provide differentiation feel they need to insource to protect their intellectual property.

Despite the advantages that insourcing may have, a growing majority of OEMs see outsourcing as the best way for reducing development costs while still being able to take advantage of new technologies. A prime example might be the new 4^th generation of i7 processors being rolled out by Intel. If an OEM was going to directly embed them, they would need to gain expertise and knowledge of the new Z87 Express chipset. It also might be difficult for an OEM to even obtain those components because they are likely to be in tight supply and the Intel alliance partners would be in a position to have better access. For similar reasons, an OEM would be better off looking to an embedded supplier for an AMD G-Series SoC based hardware product. In that case, the OEM could be focusing their internal resources on activities to develop an HMI application that fully utilized the embedded graphics capability. Doing so could make their product sparkle in ways a competitor might find difficult to duplicate let alone surpass.