Six months ago, Lynx Real-Time Systems -- well known for its LynxOS proprietary real-time operating system (RTOS) -- unveiled a courageous strategy to embrace Linux, rather than face it in armed combat on the embedded battlefield (see: Lynx+Linux -- a Dual OS Strategy).In what must have demanded a strong blend of soul searching, strategic vision, and self confidence, Lynx Real-Time Systems embarked on a dual-pronged plan of remaking itself in the image of the Great Penguin: (1) create a branded distribution of embedded Linux called "BlueCat Linux", absent a real-time emphasis; and (2) build binary Linux application compatibility into the company's mainstream (and well respected) RTOS product, LynxOS. After all, Lynx Real-Time Systems had built its business on being UNIX-compatible. "Why not," they probably reasoned, "become 'Linux-like,' and hitch a ride on the Linux popularity rocket?" While other proprietary RTOS vendors were busy defining their differences from Linux and arguing why they offered a better solution, Lynx Real-Time Systems was positioning itself to become one of the strongest supporters of Linux in the embedded market. Rather than putting effort into training their sales, support, and technical staff to sell against Linux, Lynx began preparing themselves to support Linux for customers who wanted it -- while no doubt hoping to continue supplying LynxOS whenever possible.Was it a bait and switch plan? That's tough to ascertain from out here. But one thing's for sure: after months of promoting BlueCat Linux on billboards alongside Silicon Valley's busy Highway 101, and countless full page magazine ads depicting slinky blue-faced Ms. Cat romantically sharing drinks with yellow-nosed Mr. Penguin, Lynx Real-Time Systems has passed far beyond the point of no return.Asked if his company's Linux strategy is bearing fruit, Lynx Real-Time Systems chairman Inder Singh says "after only two months of shipping BlueCat Linux, we already have some very major design wins, two of which are potentially millions of copies." Singh explains that his company is now able to support its customers with a hybrid model, based on a choice between standard Linux (BlueCat Linux) and a Linux-like RTOS (LynxOS). According to Singh, some customers who are initially interested in LynxOS move to BlueCat Linux, some go the other way, and some actually decide to use both. The result of this dual-OS strategy, says Singh, is the ability to meet the needs of more applications and, consequently, more customers.The embedded market's zigzag towards, away from, back to . . . UNIXRoughly fifteen years ago, Lynx Real-Time Systems was one of a handful of companies that had begun to promote UNIX-derived and UNIX-like operating systems into the non-desktop/non-server embedded market -- which by many accounts absorbs more than 90% of the microprocessors produced each year. VenturCom's VENIX and HP's HP-UX, among others, were examples of products that were based on AT&T-licensed UNIX source; true UNIX, but with real-time performance enhancements added. Two companies, Lynx Real-Time Systems and QNX Software Systems, were among those taking an alternate approach. Each created independent, "clean room" RTOS implementations that boasted command line and application program interface (API) compatibility with UNIX -- but with the performance, flexibility, modularity, and other constraints required by embedded applications in the industrial, military, aerospace, and commercial markets. But before the fledgling UNIX-like RTOS market could gain much of a foothold in the embedded applications arena, a new phenomenon known as "IBM PC compatibility" -- and a new operating system known as MS-DOS (a.k.a. PC DOS) -- spilled out of the desktop market into the embedded space. The result was an increase in fragmentation among embedded operating systems -- with DOS, Windows, and dozens of proprietary RTOSes fighting each other for market share, and "home grown" customized implementations remaining in the majority. Now, fifteen years later, two factors have conspired to alter this status quo.
First: Moore's Law. The growing complexities of system components have increased dramatically as silicon has become increasingly dense and functionality has skyrocketed. The result of this cascade of technical complexity is that only a few operating systems -- and no home-grown efforts -- can any longer meet the challenge of keeping up with the latest technologies.
Second: Linux. Linux represents an open-source operating system that supports a set of well known UNIX standards, while offering a level of modularity, flexibility, and functionality that makes it suitable to a wide range of embedded applications.Linux has been called a "disruptive technology." Certainly, its impact on the embedded software market has been profound, forcing a reexamination of fundamental assumptions. Lynx Real-Time Systems and QNX Software Systems, two companies that sell UNIX-like RTOSes have both taken the approach of recasting their "UNIX-like" products as "Linux-like." Interestingly, both of these companies are founding members of the newly formed Embedded Linux Consortium (www.embedded-linux.org). Of the two, Lynx took an earlier and significantly more pro-Linux stance when, last November, the company announced plans to offer its own "BlueCat Linux" distribution and a future version of the LynxOS RTOS that would run unmodified Linux application programs. By contrast, QNX recently unveiled a more modest move in the Linux direction (see: QNX goes "Linux-like").A dramatic gestureToday, Lynx Real-Time Systems takes another step along the path of transmogrification -- by adopting a new corporate name. Let's see . . . it's a "LynxOS" company . . . but it's also a "Linux" company . . . it's a "Lynux" company! It's LynuxWorks, Inc., to be precise.Congratulations, LynuxWorks, on your new name!
http://www.linuxdevices.com/news/NS2256662149.html
The Way of the great learning involves manifesting virtue, renovating the people, and abiding by the highest good.
没有评论:
发表评论