In the August 2004 IBM Power Architecture Community Newsletter Genesi’s Sven Luther and Bill Buck describe the Avalanche Desktop Reference Management Platform where Genesi is currently working on together with Avalanche.
The Avalanche Desktop Reference Management Platform
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
i think this is very good news! the pegasos is the best non apple PPC implementation on the market. cant wait to see how this all works out…
That’s excellent news for all Pegasos users & fans (count me in;-). If Pegasos will be made in larger quantites, price will drop, more people will get computer able to run MorphOS, so maybe someone will port new applications to it. As they say, MorphOS is Trojan Horse of Pegasos. You get nice mobo, put Debian on it, and then you’re getting curiouse about MorphOS cd that came with board. And then.. ;->
BTW, it’s the first time when I put a OS related newsiteam on my page BEFORE OSNews did. 🙂 Those OSN editors are quick. ;-D
…..i wouldnt compare anything to a “trojan horse” if you want people to like it….as nothing with the words “trojan horse” is ever good
….with this one exception
This is great news!
Hehe, maybe you’re right. ;->
i read this and wasnt very moved at all.
prep… stillborn. died a failure.
chrp… stillborn. died a failure.
all this Avalanche is, nothing more than CHRP with Linux installed. can we all go oooh eeeerr…
yes! amazing. a CHRP platform from 1997.. with Linux installed..
Unfortunatly I am not all that blown away. A ‘desktop reference’ guide. Who exactly is this aimed at?
I also fail to see why Genesi is part of this, unless they have decided to bunk morphos and concentrate on selling overpriced linux PPC boxes?
,,overpriced linux PPC boxes?”
@ceaser
I also fail to see why Genesi is part of this, unless they have decided to bunk morphos and concentrate on selling overpriced linux PPC boxes?
You’re gonna sell PCs with AmigaOSes to whom?
,,overpriced linux PPC boxes?”
Could you show me another PowerPC board with same set of features and lower price? And what’s wrong with CHRP platform?
If im going to run linux, does it matter what it runs on?
genesi sell pegasos II boards (quite expensive for a motherboard with cpu) and morphos right? now they have partnered up and we see this ibm article and a page with little information at https://www.avalanchecorporatetechnology.net
who is this targetd at? how will this sell pegasos ii boards?? its another spin on linux standardisation integrated with CHRP hardware.
it is expensive beyond belief to think of using a pegasos ii board as a thin client as in the example.
who is going to buy this? is a company going to go mmmm avalanche eh? lets buy 5000!
PREP/CHRP *NEVER* took off. Did hardware manufacturers and OEMS flock to it? who sells chrp systems? genesi aaand?
ibm has been trying to push CHRP for years. nobody cares about it. genesi pop up and build a ppc mobo. ibm says Yeah! we can sell our fabbed supper G* processors to another customer.
chrp didnt set the world on fire.
businesses didnt buy into chrp.
genesi boards are expensive and uber niche market.
whats the difference from running KDE on x86 to running KDE on pegasos ii board? nothing.
so who/what business is going to crossgrade to linux on a ppc board? when for far cheaper they can move to linux on probably their existing x86 base without buying new hardware?
all the article spoke of was using debconf and doing some package upgrading. oooooooooooh!! amazing!!
what WAS the point of the article???
and yes, i do consider €1,125.00 WITHOUT VAT an expensive linux ppc system.
realistlically to move this in any volume, it needs to be half to 2/3 the price. €800 euro. that will never happen.
i love ppc, i make my living on iSeries/AS400’s. that is a VERY sweet ppc architecture.
morphos is a intersting jewel in the genesi crown. but no mention of it anywhere in the article. its all linux/debian.
if i was trying to sell pegasos II boards to oems or that, I wouldnt be using debian to do it.
now if the article had been based around morphos and they integrated that into the ‘Avalanche Desktop Reference Management Platform’. Id be more inclined to be impressed.
however Im not. It reads more like someone just did a network administrators course with some corporate lan management/desktop management thrown in.
The “Avalanche” server can be established any where. We think the connectivity provided from a low cost complete solution could be very interesting when teamed with the complete open source IT infrastructure. What we would like to do is figure out a way for an organization in each country to mirror the Avalanche Pilot in their country and in their language. There are complete open source solutions to manage the localization in Debian and Gentoo (or something very different and very special with MorphOS). The idea is to sell the Pegasos like a cell phone and manage the service of maintaining the machine which generates monthly revenue. That is was the DRMP is all about. The results are selling and will sell the platform. The Pilot is about proving what is possible. With the validation given by Avalanche, which is in itself a consortium of “big” companies, other “big” companies will feel better about trying and migrating to a complete new system solution. The UserLinux article/effort mentioned here on OSNews before still needs a little bit more focus for big companies, BUT it is very much in the right direction.
The objective is to validate a complete OSS stack from OS to plug-in and the efficiency of PowerPC in ONE effort. Less power, less heat, greater reliability is more than just PowerPC marketing. The PegasosPPC *sold in this way* has NO moving parts, does not need a fan and in a thin client configuration fits in a shoe box that can be bolted under a desk.
Look at this picture closely:
http://www.pegasosppc.com/image.php?id=1170
The Pegasos doesn’t need that 500w power supply either. That is a Pegasos thin client running for an internet cafe and done intentionally to show how little space it needs in a standard case. In the meanwhile, we are selling the ODW as described. You can buy one on the PegasosPPC site. The price will can down significantly when we have everything in place.
http://www.pegasosppc.com/odw.php
This comes with not only the tool and development features, but we have already updated to the 2.6.8 kernel that was just released this past Saturday evening, so it stays CURRENT. We do this as described in the IBM article.
Once we perfect this we will sell the package of the system and the service on an annual basis which includes hardware upgrades (remember we can swap CPU cards from the motherboard).
We could also sell the ODW with a Zend package, a MySQL package, etc. We will have many things go on sale on the Freescale site hopefully before the end of this month if everything stays on track.
Please look at this slide (in fact look at the whole presentation! 🙂 ):
http://www.genesi.lu/olp.php?slide=8
The Pegasos as it is today is shown in the upper left. IT IS A PROOF OF CONCEPT AND A DEVELOPMENT PLATFORM. This allows software development to occur and the system can be deployed *as it is* in the next slide:
http://www.genesi.lu/olp.php?slide=9
What is said on the next slide is exactly what we intended to explain.
http://www.genesi.lu/olp.php?slide=10
We never said anything about a Pegasos PDA (eclipsis) for example with the form factor of today’s board, *BUT* P2P video/voice streaming developed on a Pegasos today can be of *great* benefit to a new smaller handheld design with new CPU features that could included in future designs. Another example: while the Pegasos is not in a blade form factor YET a cluster of Pegasos machines can be built together and it can function as such NOW. The primary means of node to node communications in the market today work with 1:1 extensibility. This is being done now and we hope to be making an installation as part of the Avalanche effort that WILL demonstrate this. We are a systems integrator and being positioned in this way lends itself to these kind of possibilities — especially if we have everything ready and waiting when the new handheld, cluster or whatever first boots.
This is what has to be kept in mind:
http://www.genesi.lu/olp.php?slide=11
We have access and control to every element in the technology offering. As an example, SIM and smart cards are common tools used to facilitate payment or security. If taken metaphorically as threads of different colors/consistencies they can be woven into a fabric that is more durable than a thread itself. There is no one item that provides the solution, but rather a number of items that can be combined together in a unique manner and herein we can find a solution. The opportunity we have is not bound by one technology, such as a smart card or SIM, we can integrate single items such as smart cards with other technology we can control to create something unique. For example, a particular SIM can be linked into the OF or even linked to the hardware, CPU or OS distribution itself. The ability to manage all these pieces makes the opportunity more compelling for amny situations.
The is a larger discussion looming. This is related to the security vs. privacy discussion. We see security as a technical standard and privacy as a matter of personal choice. In the same way that you provide information about yourself to enter into a mobile telephone contract you would do so here under this scenario. You exchange financial accountability for the convenience of calling from nearly wherever you want and a free phone. That is the trade-off and that is the point — the user made the decision to do it. This does not happen in the same way when you buy a computer, as there is no billing accountability and no direct tie into the connectivity which provides the link to the Internet. It has been proven and will be difficult in a post-Napster world to change that. There is and will be ID theft, fraudulent credit card transactions and issues of piracy of intellectual property that threaten the very foundation of the internet. This will not change until there is an alternative that does what people want it to do, better, more conveniently and….legally. We predict that this cannot happen until technology itself becomes focused on providing a better experience rather than trying to prevent one.
The Avalanche project itself provides the framework for more than the management of the Desktop. It is in fact the same system required for the distribution of digital data. Each Avalanche member is being upgraded when they boot the machine and this is what IS attractive. Someone IS managing the updates, the latest and greatest so to speak. The hardware for Avalanche is standardized on the Pegasos. This solves another problem of a multitude of various hardware configurations and distributions of Linux that all need their own tweaks to work in the same way as another. The means to make the offering would be the same for a project of much greater breath and depth. We must walk before we run, but we are in the right direction.
You will understand how we could also support a large scale deployment as required in the same way – Avalanche is for a Desktop reference effort, but in fact (to repeat) it is a digital distribution system. The same configuration could be used to support online a complete enterprise Linux solution based on one OS, etc. (it could also be a television/video game distribution system using MorphOS). Think about this: to get a fax you had to have one. If you have something to offer that is valuable and make things work better people will buy it.
Hope to hear from any of you who want to respond! 😀
Sincerely,
Raquel and Bill
bbrv@
I think that is a very smart idea. Personally I think that the Avalanche concept will can be addressed to a lot of other markets sucessfully…
Is to get the price down.
That currently, a Pegasos II board with a G4 900mhz CPU costs twice what a high end Athlon 64 3200+ Board and CPU cost, doesn’t cut it.
I can understand a slight premium…
But a price that’s multiples of the competition, is going to guarantee that the Pegasos Board is only bought by zealots.
If someone were to bring out an ATX Logic board that used the wealth of Mac Compatible processor chips out there…
Then, I could build a system fairly cheaply.
I’d be OK with paying $150 for a bare logic board, buying my own G4 CPU for $250 and STILL coming out cheaper than the current Pegasos OR AmigaONE Offering.
But, I’m not going to build a system to “play” with for $1500.00, when for that same money I could build two much more powerful systems.
Microcode Solutions was working on a PCI Card that would let you drop a PPC CPU into existing PC’s.
If that were brought to market, with MorphOS…
You’d have a good solution for people who want to “play” with an alternative OS.
One would think the point of being in business is to sell stuff…
Not to price yourselves out of a market due to pride and ego.
@Al Hartman:
.Hundreds millions of PC are sold each year,
.More than 2 Millions of Mac are sold each year,
.Several (hopefully) thousand of Pegasos have been sold so far
Pegasos (& Mac) will never be as cheap as PC hardware. You cannot compare them…
And btw, nobody is asking you to buy a Pegasos to “play” with MorphOS. Just like nobody would pay for a complete Mac just to play with MacOSX…
Leo.
I understand that PPC platforms in general cost more, and that’s understandable to a degree, but the point still stands that a 686 will take you further on less cash. It has to be a priority to lower the cost gap. Period.
Say I have $1000 to spend on a system. I really want a Pegasos, but what type of a Pegasos will that money buy me? What kind of a PC?
I really want to see Pegasos do well, though…
@Leo
So you’re saying Biostar or FIC or ANY logic board maker makes Millions of boards?
If FIC can produce a logic board for x86/AMD Chips for under $100, are you saying it’s impossible for a company to produce a PPC based logic board using most of the same chips and design for $150?
I find that hard to believe.
And until someone does that, there NEVER will be economies of scale to lower the price.
Instead of spending money on advertising boards, just get them out there cheaper…
I can’t believe Apple can sell eMacs for $800 for a complete system, and IBM can’t make and sell a PPC ATX motherboard and chip for under $900.
Motorola and IBM used to make CHRP boards. And they didn’t cost that kind of money.
Motorola used to EASILY undersell Apple on it’s StarMax systems with a much more powerful chip.
That people keep buying this story that PPC Motherboards are expensive because they aren’t made in the same quantity as PC logic boards is B.S.
There isn’t ONE brand of logic board in the PC space that gets made in the millions.
And if someone invested in a run of Logic boards that took a PPC processor and used PC Standard hardware, they’d EASILY sell out a run of 50,000…
Even as over priced as they are, the Pegasus II and AmigaONE boards sell through VERY quickly. They can barely keep up with the demand as it is.
And that’s ANOTHER reason the price is so high…
There’s still a market for the systems at the high price. Until initial demand drops off, and it stops being a ‘Sellers Market’, the prices will stay artificially high.
>If FIC can produce a logic board for x86/AMD Chips for under $100, are you saying it’s impossible for a company to produce a PPC based logic board using most of the same chips and design for $150?
>I find that hard to believe.
Sure a company *willing* and having enough ressources could easily do that. You only need to find one Company which follows these 2 needs…
=> But then, where would they find the millions of customers ?
What PPC OS, except MacOS, can sell millions of PPC-boards ? I don’t see any.
Leo.
to Novells ZENWorks ?
Who should take avalache if an industrial strength solution is avaiable by Novell (SuSE, Ximian) ?
Cheers,
Frank
Being able to produce that much boards is not the problem. Selling them is the problem. And for that you need to find the people and then to convince them to buy it…
Leo.
> to Novells ZENWorks ?
Is this a project you worked on or work with?
@Darth_X
is Avalache GPL or BSD ?
So, whats the point ?
Cheers Frank
Which is better, working on a GPL or BSD project for free, or working on a project where you get paid when the project is finished and shipping?
Thanks Darth_X, you get it.
Avalanche is a coop. It is a different “business” model.
The objective of DRMP is to lay a opportunity pipeline between the member companies and the developers themselves. Novell leverages open source to sell their stuff. In this case, Avalanche is providing a service to its members by financing and validating solutions (on behalf of its members) that can be contracted as needed from developers themselves. The desktop is just a starting point. There will be many customization requirements and Avalanche will have a progressively useful resource to use — not sell.
R&B 🙂
i dont know where you get the price that you all think a complete system will cost. I spent $900 for a complete system. If you look arround you can build one for a great value.
Pegasos II G4 at 1000MHZ not 900!!
Radeon 7000 AGP 64MB version
ATX Mid Tower case
CD/CD-RW DVD combo drive
Corsair Twin-X 512MB DDR
Total price $900 total all new parts.
whats better ?
using a system from a company which no one knows how and how long they will survive, or using a Software from a company that shows that in can survice (ie. Novell) ?
Thats the point of view of a user, not of a developer.
As a user, if you don’t know if you partner can survice, you possibly want to have the source and rights on it.
BSD and GPL are also nice for this.
I guess it the same “different Business model” as the Netintrusion detection System had based on OpenBSD on the Pegasos.
Cheers Frank
Frank, as the operator of this site — http://www.ppcnux-projekt.de/ — and as Genesi is a former advertiser, we do not really want to get into a argument with you here. Let’s call it a discussion as we have had before. Example:
http://www.ppczone.org/modules/newbb/viewtopic.php?topic_id=20&foru…
There is a bigger issue than “survival” (we have demonstrated we can do that). Will (as asked here — http://penguinppc.org/ ) the PowerPC find a role on the corporate desktop? This seems to be the issue of the survival of Genesi and whether or not communities like Debian get interested in the platform. For us, IBM and Freescale are the most logical to enlist as supporters (is there any one else making the PowerPC?). This thread was focused on an IBM article. Freescale is also providing support and resources.
http://www.freescale.com/webapp/sps/site/overview.jsp?nodeId=018rH3…
(see Genesi featured in the middle of the page and link)
They have also made us a Gold Sponsor at these future events:
http://www.freescale.com/webapp/sps/site/homepage.jsp?nodeId=02M0yl…
http://www.freescale.com/webapp/sps/site/homepage.jsp?nodeId=052577…
Our role is too provide a file server/thin client based Internet Cafe, as we did here:
http://www.pegasosppc.com/image.php?id=1170
The Pegasos does not need that huge power supply, but it illustrates a good point. There is alot of empty space in that standard case.
As for the OpenBSD fiasco, there is plenty online about this. To bring it up again had really nothing to do with the article or even the Avalanche “business model” which is to provide a service to it members. There is a *significant* distinction between packaging free software with expensive commercial applications and packaging free software and in conjunction with developers customizing that free software in exchange for money to support Avalanche member needs.
Did you read this article that was featured here on OSNews?
http://www.desktoplinux.com/articles/AT5016155846.html
This is a great project and a step from the other direction toward what we are trying to facilitate through Avalanche (that has chosen to use our product).
Hope to see you at SNDF Europe.
Raquel and Bill
[email protected]
We would like to invite all interested PowerPC Developers to Smart Networks Developer Forum in Frankfurt on 30 September. You can attend the Tech Show in the morning and attend Workshops. In the afternoon we will have some special PegasosPPC oriented presentations, including a series of workshops on AltiVec. Lunch is included and developers who have been pre-approved and attend will be given an Open Desktop Workstation compliments of Freescale and Genesi. 🙂 In exchange, developers are ask to file a monthly report on what they are doing with the machine – only basic guidelines for development will be given.
Please contact us if you are interested in developing for the platform and attending PegasosPPC Day at SNDF Europe.
Â
Raquel and Bill Â
[email protected]
Genesi, when will the next developer show be in the united states please? Hold it in Chicago if you can please.