Novell to Launch Quick-Response Linux

Krishna Dagli writes to mention a C|Net article about Novell's upcoming real-time Suse Linux Enterprise product. From the article: "Real-time operating systems can respond to external events within a guaranteed time frame, a feature that mainstream business computing doesn't generally require but that's necessary for some areas, such as aircraft radar. But in a move that indicates the flexibility of Linux, Novell plans to begin selling the real-time variant of the open-source operating system next month."

Quick Response and Real-Time

[postmortem]

...are not related. Real-time operating systems have shadulers that sort tasks per various priorities. Which mean that task admitted with low priority isn't going to have quick response.

Re:from the that's-awful-fast dept.

[Lisandro]

It's actually not a matter of speed. A realtime OS [wikipedia.org] guarantees that a given external event will trigger a response in no more than x milliseconds. This might be slower (or faster) than a non-realtime OS, but it's not the point. There are a lot of applications where things must be attended within a given timeframe (radars, as mentioned, industrial automation, nuclear plant safety devices, etc.). Achieving this this is not a trivial excersice.

This is not about the response time after you click on your Office document - and in fact, the (huge) added complexity of a RTOS is nor needed nor desired in most desktop and even enterprise uses.

Re:from the that's-awful-fast dept.

[Doctor Memory]

WTF? "real-timeness adds overhead"? The whole point of RTOSes is to remove the overhead and turn almost everything over to the application. Lots of RTOSes have only basic memory management and I/O is largely limited to device discovery and initialization -- everything else is left to the application, because only the application can define how long an operation may take. Some telephone switches can even be rebooted while the applications run!

You might want to take a couple of computer classes before you post here again.

RTOS overhead...

[Junta]

A characteristic common of multi-tasking RTOSs is that timeslices are very small, and context switches extremely frequent, to help assure that apps will be able to do something in the required timeline, which is not necessarily a small amount of time, but in many applications it is a short time. There are more complexities, but the bulk of the 'overhead' mentioned in the GP is probably the high percentage of time spent doing context switches rather than useful work. The key word here is multi-tasking RTOSs, you describe cases in which the OS is largely a base for a single application that does everything itself. This is very common in the RTOS world, but their do exist areas where pre-emptive multi-tasking RTOSes fulfill a need....

Of course, your point is correct too, a requisite step with respect to this particular aspect of a RTOS is generally considered to be doing whatever it takes to make the context switch as fast as possible. In a server or desktop OS historically context switches aren't nearly as optimized as the solution can be for most applications to lengthen the timeslice. Server platforms particularly tend to have long timeslices as their activity isn't highly interactive, and the percentage of the time the system can do useful work is more important than each application being able to do their thing in a timely manner. On a desktop OS, timeslices being shorter is helpful (fairly recently a lot of debate and configurability happened in linux with respect to this), as multiple applications are more directly interacting with users. With a handful of processes it isn't really noticeable but as it scales up the delays get into perceptible fractions of a second. Having a smooth interface is a bit more along the lines of a design point of RTOS, so it makes sense. Anyway, there is a penalty paid, but the timeslices for satisfying human interaction requirements may still be orders of magnitutde higher than those required for RTOS applications, depending on what's going on.

Re:Sounds like it might be

[paulbd]

Ingo Molnar and Andrew Morton's original 2.4 patches for better soft-real-time performance were prompted by complaints from those of us in the linux audio (pro-audio/music) software community. Systems like JACK [jackaudio.org] and applications like Ardour [ardour.org] were written around this kind of functionality.

Ingo has picked up the ball and run with it for 2.6 to the point where a kernel patched with his most recent patches is basically hard-time except for cases where a badly written device driver screws it up. And increasingly parts of his work have become part of the vanilla kernel so that even a 2.6 kernel now works pretty well unless you need very low latency (as require for real-time monitoring and FX processing.

Example uses..

[Anonymous Coward]

Passur [passur.com] makes a system system uses Linux with RT patches on the back end for the processing of radar data. About 6 years ago, I helped install and perform low level maintenance for some of the Passur subsystems at IAD for a major carrier there. I also met with and worked with the Passur engineering team near Islip airport in NY and got a day or two of technical training from them. Disclaimer: I have not worked there in over 5 years so my information may be a little out dated.

The Passur system basically piggy backs off of the airports radar system but is in no way actually connected or related to the airports system. It is 100% passive (hence the name Passur). The unit is "synchronized" and tracks with the airports radar and reads the echoes and bounces from the planes, including the IFF [wikipedia.org] data from the planes. I believe at IAD, the Passur antenna array was at least a mile away from the airports radar too. This data is collated and sent to the airlines or airports tracking system for real time analysis of where planes are and how long until they get to some other point. The data is also sent to Passur for collection for resale to other airline related data companies or smaller airlines that do not have their own systems. The more airports and airlines they have on board using their system, the more data they have and the more useful the data is. Only one system needs to be in place to provide useful data to a much larger area. An example is the DC area, IAD radar can actually serve IAD, DCA, and probably BWI as well.
When the customer service agent at a gate announces that flight 123 will be arriving in 15 minutes and the monitors around the airport that show flight arrival times are getting that data from a Passur or similar system (or they just make up the times to shut people up).
You can obviously see that this level of synchronization requires some RT functionality and a lot of initial tuning for the timing.. The Passur system was using this back in 2000 and I have no idea what the actual state of RT Linux was back then but obviously it worked good enough. I believe the main machine was Redhat based.

One of the advanced functions of this data collection is the ability to record and "play back" what was going on in the air at a certain time as well. Their Islip facility captured the path and data from TWA flight 800 up to the point that it disappeared. I saw that played back when I was there.