|
|
 |
|
Viewing report
|
|
|
|
2009 Embedded Development Tools & Real-time Operating Systems
Embedded Market Forecasters, July 2009, Pages: 47+
The analyses of evolutionary forces following Darwin shows that change, when it occurs, is rapid and decisive thereby changing irrevocably the landscape and the survivors. We have witnessed this in technological terms every decade since the 1970’s, and we are currently witnessing such an irrevocable upheaval in the marketplace for software design and development.
The economics of the embedded industry are changing. Companies are no longer buying based on claims of technological advantage – CEOs and CFOs are insisting on seeing a long term ROI before changing their processes and investing in other tools.
Considerations once the domain of IT or large systems developments are finding their way into traditional embedded markets. Model Driven Development (MDD), particularly modeling that provides code reuse, autocode generation and re-porting of APIs with changes in underlying hardware, has been shown to provide a significant ROI over non-MDD comparable developments. Expect this trend to continue.
Post-shipment support is becoming a financial drain for many organizations. EMF data continues, year-over-year to show that non-commercial (roll your own – RYO) Linux and communications middleware fall far short of commercial tools. In time, RYO developers will face an economic crisis – they can’t be scaled efficiently.
Look to Requirements Management (RQM) and Change Management (CM) to become a broadly accepted “best practice”.
In additional Validation and Verification tools will become required for many embedded applications.
The principal expanding market that can drive the embedded industry to new heights involves devices that can communicate with IT-based databases, thereby enabling a broad range of what are referred to as “smart devices”. Developers of these device/systems will require more than RQM, CM and MDD to fully integrate the necessary IT technologies.
This is the future – and it is near-term.
Forget about DSO and safety critical OSes. Such RTOSes as Nucleus, ThreadX, Monta Vista Linux and Micrium offer an important alternative for many smart device applications. With tens of millions of these OSes successfully deployed, the idea that DO 178B or MILS certified OSes should be used for all applications is sheer folly. Choose the OS and chip set that best serves your application.
Device Software Optimization (DSO), although an important step in design and device efficiencies (particularly since many DSO tools are OS agnostic) is not the final step for design and device enhancement. The growth of simulation-modeling tools and the emergence of UML as the preferential tool of choice (permitting legacy upgrades, software component reuse and automatic code generation) have closed the design loop and have made DSO use limited.
Today, military and government markets are moving cautiously but decisively towards a software centric purchasing model. Interestingly, as telecom and other communication markets for merchant computer boards decline precipitously, the value of software as a product offering has increased significantly.
Also, embedded vendors (as well as developers) should be aware that new developments in software product lines (SPL) can be used. The goal of SPL is to reduce the time, cost and effort required to create, deploy and maintain similar products. To achieve this goal, the ultimate solution must minimize duplicate effort, maximize commonality among design and implementation assets, and optimize reuse of effort across similar products within each of its product lines.
Given benefits of SPL as a basic software solution, the next question is how does one also support variations in requirements and/or test? The answer is that SPL is really a piece of a larger process called Product Line Engineering (PLE).
A PLE engineering lifecycle framework offers a consistent and integrated solution for all system and software tools and for the reuse of all systems and software assets, across the entire product development and maintenance lifecycle. This is much more effective than traditional development approaches. Essentially this takes the same approach to commonality and variability management in SPL and applies to other tools like DOORS for requirements management. This means one can use a common set of features and variations to manage requirements and modeling elements concurrently.
IBM/Rational has raised the bar to a new level of sophistication that includes systems and systems-within-systems development, and systems deployment and support. The intrinsic ability to reintegrate legacy code into new software designs and the concomitant ability to carry along detail documentation solves the problem of what happens decades from now when the original developers have long left the scene. We should see adoption of these capabilities for military developments that have life spans of many decades.
Ready or not, this is the future and embedded vendors should pay heed.
The status quo is no longer with us – who remembers DEC, COMPAQ, and Hayes Modem? Who will be the next to fall?
Customers who bought this item also bought
Embedded Systems Development Survey 2007
2009 Embedded Hardware/Software Design Preferences: Design Starts, Completions, Cancellations, Processor/Software Utilization
2009 EMF Embedded Executive Dashboard
Operating Systems Used in Embedded Systems - Voume 5: Mobile Software Stack
2007 EMF Embedded Executive Dashboard
Operating Systems Used in Embedded Systems - Voume 4: Multi-Core Components and Tools
Embedded Systems: A Contemporary Design Tool
Operating Systems Used in Embedded Systems - Volume 1: Linux
Operating Systems Used in Embedded Systems - Volume 3: Embedded/Real-Time and Mobile OS
Operating Systems Used in Embedded Systems - Volume 2: Windows Embedded and Mobile
Embedded Processors: Global Market Demand Analysis
2009 Embedded Development Tools and RTOSes: World Markets, User Perspectives and Strategic Issues
|
|
|
|
