2013年3月25日 星期一

The open source software movement has forever transformed the mobile device landscape

Single board computer, Panel PC, networking appliance

The speed of innovation in automotive IVI is making a lot of heads turn. No question, Linux OS and Android are the engines for change.

The open source software movement has forever transformed the mobile device landscape. Consumers are able to do things today that 10 years ago were unimaginable. Just when smartphone and tablet users are comfortable using their devices in their daily lives, another industry is about to be transformed. The technology enabled by open source in this industry might be even more impressive than what we’ve just experienced in the smartphone industry.

The industry is automotive, and already open source software has made significant inroads in how both driver and passenger interact within the automobile. Open source stalwarts Linuxand Google are making significant contributions not only in the user/driver experience, but also in safety-critical operations, vehicle-to-vehicle communications, and automobile-to-cloud interactions.

Initially, automotive OEMs turned to open source to keep costs down and open up the supply chain. In the past, Tier 1 suppliers and developers of In-Vehicle Infotainment (IVI) systems would treat an infotainment center as a “black box,” comprised mostly of proprietary software components and dedicated hardware. The OEM was not allowed to probe inside, and had no ability to “mix and match” the component parts. The results were sometimes subquality systems in which the automotive OEM had no say, and no ability to maintain. With the advent of open source, developers are now not only empowered to cut software development costs, but they also have control of the IVI system they want to design for a specified niche. Open source software, primarily Linux and to some extent Android, comprises open and “free” software operating platforms or systems. What makes Linux so special are the many communities of dedicated developers around the world constantly updating the Linux kernel. While there are many Linux versions, owned by a range of open source communities and commercial organizations, Android is owned and managed exclusively by Google.
Linux has further opened up the possibilities with safety-critical operations and multimedia communications. Hardware companies have followed suit with more IVI functions built onto a single piece of silicon, improving security and performance.
The available power of multicoreSoC hardware hosting a Linux operating system is fueling rapid expansion in vehicle software in the area of telematics. In Europe, for example, by 2015, all new cars must be equipped with the eCall system, to automatically alert emergency services in the event of a collision. Services such as insurance tracking, stolen vehicle notification, real-time cloud data (traffic, weather, road conditions ahead), car-to-car communication, driverless car, diagnostics, and servicing are also made available via in-car Internet services. To operate in this space, IVI hardware needs to have multicore processor support, GPU/high-performance graphics with multiple video outputs, Internet connectivity, and compatibility with existing in-car networks such as CAN, MOST, and AVB. Several components are already on the market, and the future potential is exciting.
Consolidating multiple functions into a single Linux-based Electronic Control Unit (ECU) allows for a reduction in component count, thereby reducing overall vehicle costs. Maintenance becomes easier. And the wire harness costs are reduced as the total ECU count drops. As Linux becomes more widespread in vehicles, additional technologies will consolidate – for example, instrument clusters and AUTOSAR-based ECUs may coexist with infotainment stacks. It’s also important to realize that the complexity of software and the amount of software code used will only increase as these new technologies become standard. Already more than 100 million lines of code are used in the infotainment system of the S-Class Mercedes-Benz and according to Linuxinsider.com, and that number is projected to triple by 2015 (Figure 1).

Single board computer, Panel PC, networking appliance

Figure 1: Software complexity in IVI systems continues to grow. Today, the IVI system of an S-Class Mercedes has 100m lines of code. By 2015, it is expected to be 300m. A Linux-based solution, capable of scaling to handle the complexity, is mandatory.

Android apps hit the road
The Android operating system, on the other hand, was designed from the start to support mobile devices and has proved that it can serve more than mobile phones. Using the Android OS for in-vehicle entertainment provides all the entertainment features offered by a top-of-the-range, in-dash infotainment system with the addition of informative, driver-assisting content including hands-free calling, multimedia center, and a navigation system/Google maps. For an open source expandable system (whereby the framework can be extended and applications can be developed for it), the Android OS can be enhanced to support multiple audio and video feeds. For example, IVI audio requirements include music, phone calls, sensor warnings, and navigation announcements, which must be managed and prioritized. Managing multiple displays, with an information-focus for the driver and entertainment-focus for passengers, is also a requirement. The UI for the driver should be arranged to minimize distraction, while passengers will want as much content as possible from their UIs. But many automotive OEMs and developers ask, “Why not just use the Android smartphone and tie it into a vehicle’s dash?” Not only would this be more cost effective for the developer, but the user would have instant familiarity with the system.
One organization promoting the use of the smartphone as an IVI in-dash system is the Car Connectivity Consortium (CCC). The CCC provides standards and recipe books for tethering a smartphone to the infotainment head unit. The CCC members implement MirrorLink (Figure 2), a technology standard for controlling a nearby smartphone from the in-car infotainment system screen or via dashboard buttons and controls. This allows familiar smartphone-hosted applications and functions to be easily accessed. CCC members include more than 80 percent of the world’s automakers, and more than 70 percent of global smartphone manufacturers. The MirrorLink technology is compatible with Mentor Embedded’s GENIVI 3.0 specification Linux base platform solution.

Single board computer, Panel PC, networking appliance

Figure 2: An example of smartphone in-dash tethering: Drivers use the same smartphone apps in the vehicle as they do on their own smartphone, which provides a great deal of familiarity.

A recent example of smartphone tethering can be found in certain subcompact models from U.S. auto manufacturer General Motors. Select Chevrolet models carry the “MyLink” in-dash infotainment system.
From both a cost and ease-of-use perspective, tethering a smartphone makes a lot of sense. But there’s another reason to consider. Some automotive manufacturers are nervous about being too dependent on Google – as Google is the sole provider and owner of the Android mobile platform. Android built into an IVI system is an 8- to 10-year commitment, and a lot can happen in that time regarding license fees or terms of use.
Linux and Android driving together?
Despite the strengths of and differences between these two popular platforms, recent embedded architecture developments now allow the Linux and Android operating systems to happily coexist. And this might be a very good thing. For example, Android can be hosted on top of Linux using Linux Container Architecture (LXC) (Figure 3). The resources, access control, and security of the Android client are managed by the host Linux operating system. For system designers concerned about the security of Android, this represents a good way to offer Android app access, and keep other system functions on a standard Linux platform. Multicore System-on-Chip (SoC) platforms make this architecture even more attractive, as there are sufficient resources for both Linux and Android domains to perform well simultaneously. The CPU resources can be shared, along with memory, graphics processing resources, and other peripherals. The output of the two domains can be recombined into a common Human Machine Interface (HMI) allowing the user to select functions from both domains.

Single board computer, Panel PC, networking appliance

Figure 3: There are several ways to include Android (Android apps) in a Linux-based IVI solution. One method, which is becoming increasingly more popular, is using Linux Container Architecture. Here, Android sits as a guest OS on top of the Linux kernel. Privileges and permissions are tightly controlled.

Exciting times ahead
Both Linux and Android are extremely versatile and powerful operating systems worthy of consideration in IVI systems. We are still in the infancy stages in what these two open source platforms can do for IVI. Now is the perfect time to starting developing or to join a consortium so that you too can reap the fruits of what IVI promises down the road.




refer:http://embedded-computing.com/articles/automotive-source-drives-innovation/


2013年3月12日 星期二

How to approach to communication processors...

Generic multicore processors have been promoted as the solution to networking communication processing. In reality, they can’t address the scalability, determinism, and ease of programming required for next-generation networking infrastructure. An asymmetric multicore approach that blends multicore processors with networking-optimized accelerator engines and C-programmable libraries meets the challenges of next-generation networks.
Achieving deterministic performance is a key requirement for network operators to ensure reliability across wide variations of traffic profiles and applications. Multicore processors can meet performance challenges when running an application on a single, dual-core, or quad-core processor. However, when scaling to eight cores or beyond, performance scaling usually degrades. There are cases where eight cores deliver no better performance than four, and 16 cores actually run slower than eight.
Networking applications tend to be data-intensive, and generic multicore processors are highly susceptible to the impacts of memory latency on performance. The nonlinearities of memory latency (Figure 1) with regard to memory load combined with the nonlinearities of processor performance relative to memory latency can lead to unpredictable and unreliable performance. The innovative approach taken by LSI to solving this problem is asymmetrical multicore processors, which combine general-purpose processors with specialized accelerators for particular data-intensive tasks, resulting in an optimal, scalable solution.

Industrial PC, gaming platform, networking appliance
Asymmetrical multicore processors improve performance predictability by combining general-purpose processors and accelerators to address the nonlinearities of memory latency.
Networking infrastructure applications tend to involve complex processing, intense memory utilization, and real-time, deterministic requirements. Asymmetric architectures address these challenges by seamlessly allocating the work between general-purpose multicore processors and specialized acceleration engines. These accelerators are specifically designed to tolerate memory latencies and perform predictably. This approach also enables the application to be built using fewer general-purpose multicore processors with far fewer lines of code. The asymmetric approach simplifies scaling challenges and delivers more deterministic performance at lower cost and power.
Networking applications demand a flexible approach to OSs. This flexibility is required not only to meet application requirements, but also to support the smooth migration of OEM legacy software and give designers the ability to choose the right OS for a particular application. It is important to simultaneously support multiple OSs on different cores without introducing overhead. At LSI, our hardware and software has been architected from the ground up with all this in mind, providing flexible support for the range of OSs used in networking applications.
refer: http://embedded-computing.com/articles/an-multicore-done-right244/