Hardware Headaches Inevitable? 73
JaneWalker6847 writes "Don Becker, co-founder of the Beowulf project, describes the inevitability of hardware administration headaches and warns users not to expect a silver bullet to solve these problems." From the article: "We're about to see another revolution, which is in network adapters -- that we [will] talk directly to [them] from application level. That's a massive change in how you interface with them. And that brings about a new round of device drivers completely unlike the device drivers we had 10 years ago. So, that part of the world isn't going to stabilize anytime soon."
Re: (Score:3, Funny)
He just stated what is. "My boss now insists that we all wear a tie to work. In a call center."
Same tone.
What really worries me (Score:5, Funny)
revolution indeed (Score:2, Interesting)
"We're about to see another revolution, which is in network adapters -- that we [will] talk directly to [them] from application level.
I hope ioctls do perhaps the same job as long as there is a module properly written to handle a specific ioctl
Or is it like controlling network firmware directly from Application ? !!!
Sounds like weird...
Imagine a malicious program kicking your Network Adapter's butt :) ...
Re:revolution indeed (Score:5, Informative)
Re: (Score:1)
Re: (Score:1, Troll)
My CPU load never goes above 5% with my integrated etherwebs running full on @100Mbps. We don't _need_ this.
Re: (Score:3, Insightful)
Re: (Score:1)
Re: (Score:1, Funny)
640KB oughta be enough for everybody!
grtz HillBilly
Re:revolution indeed (Score:4, Informative)
In a data center environment? Quite often.
Re: (Score:2)
Re: (Score:2, Interesting)
Here's a short list of companies for whom 10Gb Ethernet likely comes in handy:
And then there are systems on the lower end of the Top 500 supercomputers list.
Re:revolution indeed (Score:4, Insightful)
It seems that these old cards are on sale on the eBay and [when I look those other search results] also some are sold as new, so I guess that this is still viable technology in some places: http://www.google.com/search?q=ssl+encryption+acc
Anyway, I just have this feeling that it's not a good idea to integrate any kind of encryption technology directly to the card.
Hmm... Do I smell WinModem?
Re: (Score:3, Interesting)
Re:revolution indeed (Score:5, Insightful)
I'm not sure what that smell is, but its familiar.
Yes, TCP/IP offloaders, crypto offloaders, physics offloaders, FFT offloaders, have all existed. The only accepted offloader that has succeeded is the GPU, and that is because it was subsidized by the high end graphics people and people with game addictions. The cost/benefit of the other offloaders has not proven itself. Especially when you consider the rate of increase of the CPU speeds and the bottleneck of getting the data to the offloaded chip and back again. The Linux kernel mailinglist has been pretty much anti TCP/IP offloading because the time spent optimizing the drivers and the performance increase was often surpassed in a few months with a faster general purpose CPU and a generic driver that worked with all cards. This is also confounded when you have OS level TCP/IP stuff like ipchains or iptables that need their code in the OS and not on the card. FFT and physics offloaders have not taken off because of the cost/benefit loss when one takes into account the speed loss of getting the data onto the card and then the cost of the cache RAM to put on the card (if its ever enough), and then the time to get the data off of the card.
Now, what will make these things work?
A bus that is near or at the speed of memory bandwidth or a problem where the data does not need to go back to the host system. A GPU falls into the second half there. The display needs to know what its told, the computer does not need to know the details. TCP/IP offloaders have failed to catch on because of the price/performance benefit and their lack of ubiquity and commonality. Crypto offloaders are cool. Especially for the geek factor of having the keys stored on the card and zeroed out when tampered with. But again, the cost of writing specific drivers vs known CPU drivers for doing crypto and the rate of increase of cheap commodity CPUs over time often exceeds the cost/benefit of the crypto card.
Now, give me a fast bus and the ability to use my generic system RAM with one of these offloading cards, and things could change, but until then I expect this battle between the CPU and specific PUs to continue with no real winner.
memory bandwidth (Score:3, Interesting)
This combination would also work fine for 90% of the world's computer users, and possibly be much cheaper. Think Sempron with RAM and a minisc
Re: (Score:2)
And every once in a while you here an Intel e
RAID cards (Score:1)
Re: (Score:2)
True, but only on the high end, price has no object range.
Cheaper RAID cards are not worth using because in the event of a card failure the odds of you losing your data is still great. I've heard that some of these things cannot read drives that were setup with a different firmware revision of the same brand of card.
Again, for an in box card, I woul
Not WinModem-like at all (Score:4, Informative)
Except that WinModems are the exact OPPOSITE of the philosophy that's being espoused here with crypto offload engines, intelligent network cards, etc.
The WinModem was an attempt to take traditional modem functions and move them onto the CPU, in software. Rather than actually having a box full of circuitry that did the hardware handshaking, data compression, and all that good stuff, you just replace it with a simple device that barely connects the analog telephone line to the computer, and have the computer do all the heavy lifting.
I think the justification behind this approach is "software is cheap, hardware is expensive." Therefore, you put the 'brains' in software, and your dumb-hardware/smart-software combo is cheaper than the traditional combination of dumb-software/smart-hardware.
It's a pretty radical departure to essentially go in the opposite direction, from WinModems to these kind of "intelligent network cards," which seem more like a traditional serial modem in philosophy; they do all the work themselves and basically present the computer with a standardized data stream.
The only way that I could see this whole business being "WinModem-like" is in it being tremendously difficult to program for on non-Microsoft OSes. But that's not a consequence of the design per se, but of how I suspect MS will choose to implement it.
Re: (Score:2)
Yes, I know. I was referring to WinModems because I think that this kind of technology will fail in similar fashion (well, ok, if we look modern laptops, it can be argued that the WinModem buisiness haven't actually failed).
Re: (Score:2, Interesting)
This kind of general-purpose architec
Re: (Score:2, Interesting)
Re: (Score:3, Interesting)
I don't want yet another opaque software layer (Score:1, Troll)
in my linux system. I want open source network drivers that implement the TCP/IP stack without, say, phoning home for instance. Drivers that I could compile or hack myself if I wanted to.
Re: (Score:1)
The firmware in such a card would not even have to be upgradeable
(And don't come whining that zero bugs is impossible. MY programs have zero bugs, thanks to testing each and every possible and impossible case for everything. Takes time? Yup. Zero bugs == priceless.)
Re: (Score:1)
No piece of software, BTW, can be deemed zero-bug, unless mathematically proven to be so. And then of course there's the dimension of concurrency that adds
nil nove sub sole (Score:3, Interesting)
Re: (Score:1)
http://hardware.slashdot.org/article.pl?sid=06/08
Killer NIC anyone?
Re: (Score:2)
Things like this are already commonly done in high performance computing, where you don't want to interrupt the CPU (which is doing real work) to service message passing requests.
One example of a production system doing this is the Cray XT3 [cray.com]. You have a PowerPC 440 processor sitting on the card, along with a DMA engine. A request comes over the NIC, it will put it in the proper place in memory that you specified earlier.
Re: (Score:2)
Re: (Score:3, Informative)
Re: (Score:2, Interesting)
That being said, if you are trying to scale (think a dozen gigabit cards running at high utilization) or a significant number of high-throughput IPSec/VPN clients, then the offload hardware can really show up as a big gain. Even the OTS gigabit ethernet cards these days support of
Re:revolution indeed (Score:4, Informative)
The problem is that general-purpose cpus grow in power so quickly that the offload-engines get ever-larger problems beating them. And they get the aditional problem that they don't get packet-filtering or anything else that is not custom-written for that particular card (if it's even possible to convince the card to do it!)
It's also nothing new -- these cards have existed for literally decades, and haven't managed to make any kind of inroads, not even in specialized servers.
Have a look at this year-old Lwn-article [lwn.net] for an example listing some disadvantages.
Re: (Score:3, Interesting)
That's what this sounds like, giving the network card the kind of specialized bus and direct communication channels like the Graphics subsystem.
Re: (Score:2)
Later, it's been tried literally every 5 years. There actually *exists* "tcp offload engines", they existed 5 years ago. 10 years ago and 15 years ago. They existed in 1969 if you count the imps. (though those where external computers, not car
Re: (Score:2)
Re: (Score:1)
Re: (Score:1)
that's why it is titled Hardware Headaches Inevitable, resistance is futile...
Which gives a whole new meaning to... (Score:3, Interesting)
Ok, but seriously, maybe someone can answer me this. Why do we still need to construct massively parallel computing architectures at the platform level? Not saying we should toss the whole concept, but for the foreseeable future won't it make a lot more sense to stick with the Amazon model of chunking up into virtual machines? I know the FA says that this view is a mistake, but he doesn't explain why. Can anyone else?
Re: (Score:2)
Why? (Score:2, Interesting)
Besides, what about hardware abstraction? If we're talking directly to the network adapter, isn't this taking a step back into the past (remember when you had to hand-code ASM to talk to various video/sound cards back in the early days of PC demos and games?)
Re:Why? (Score:5, Informative)
What this new thing means is that applications send the data to winsock which hands it directly to a new kind of network card/driver which takes the data and header info and creates the TCP/UDP and IP packets on the card itself in card firmware. From there, the card wraps it up in the lower level protocols and then puts it out over the wire (or air if its wireless)
Re: (Score:2)
The person who made this ad should never EVER mention it on their resume.
Standarization (Score:3, Insightful)
The idea that your original hardware vendor can provide you with a drop in replacement for a failed card is nice, but any decent manager is going to ask you what the plan is for when that vendor goes under. You need to know that you can just buy some card off the shelf and put it in and have it work. At least with our current driver/hardware structure, you know that a 3com nic is going to work. Or if 3com goes under, you can drop an intel nic in. You may have to install the driver for the card, but it's not going to (unless you start messing with dirty cheap hardware) have compatability issues.
I guess my point here is, there's no way a bunch of companies would target big business with a product like this without there being SOME standard interface. Who wants a multimillion dollar migration to 100% proprietary hardware?
Be the first to pre-order your essential apps!! (Score:3)
Taking pre-orders now for exciting new products to be released soon!
Only $50 each!
Choose from the list below, and many more!
If you purchase 10 or more, we'll give you a COMPLIMENTARY Firewall Commander (a $100 value), to simplify the process of setting up rules for multiple applications!
</advertisement>
Seriously, why would I WANT to have to update all of my programs because a hole was found in the networking code (that they all share - because it's a full featured drop in library - BSD licenced and everything)?
Did anyone think this through? Or, is this a follow up to the "OS of the future" article?
Re: (Score:2)
Re: (Score:2)
Ultimate solutoin to driver chase (Score:1)
This guy is worth listening to (Score:4, Interesting)
It looks to me like he's telling us that drivers are not likely to go away as an issue any time soon. Too bad, but if Becker says so, he's very likely right.
Re: (Score:2)
Intelligent people can disagree.
Who do you listen to when the Linux core developers argues with the FreeBSD core team?
Being highly knowledgable in an area doesn't give you a monopoly on the truth, or the ability to accurately predict the future.
I think there was a related story in hardware (Score:1)
http://www.killernic.com/KillerNic/KillerOverview. aspx [killernic.com]
http://hardware.slashdot.org/article.pl?sid=06/08/ 08/193237&tid=230 [slashdot.org]
It seems to only Offload UDP traffic instead being a full network stack.
Re: (Score:1)
the KillerNic's pci bus will not let it hit full gig-e speeds.
Err thats not a good idea (Score:1)
From an Old skool OSI background I am astounded that someone has said this and this guy is supposed to be a freakin player not some pimply faced youth.
There are very good reasons for HAL's and Abstraction Layers.
Some people commented that its moving some of the stack onto the NIC freeing the CPU to do more useful work - but the problem is OS suppliers will still have
Re: (Score:2)
I think you miss the intended markets for those cards: hard-core gamers and high-performance streaming-media systems. Gamers in particular will pay inordinate sums to shave another millisecond off their game's ping time or get that extra FPS or two, and they don't really care how bad an idea is overall if it gets them what they want.
Re: (Score:1)
I don't see what changing your NIC will give you a worthwhile decrease in ping times.
maybe in a LAN party but I doubt they are running gigabit networks with top of the line cisco Gig switches with a 10g Core at the average LAN party.
Re: (Score:2)
The card's theory is that by moving a lot of the network stack into hardware you can decrease the latency due to OS driver/stack processing. It's a nice theory, pity that part of the latency's only 1-2% of the total latency. But then, buying magic solutions to basic problems isn't new. Think of the audiophools who spend $20/foot on deoxygenated, magnetically-aligned, high-copper-alloy low-resistance speaker cables with special xenon-impregnated insulation and amorphous-bonded iridium-plated connectors becau
He means Linux network channels (Score:2)
Nice [Editing] (Score:2)
If you're going to [will] insert words to correct someone else's writing, make sure [them] the changes are actually correct.