Slashdot Mirror
Slashback: Grids, Netscape, AMD
And Campbell's puts glass marbles in their soup pictures. Roland Piquepaille writes "We saw several grid computing announcements in the last couple of days.Of course, Gateway stole the show. In 'Gateway makes store PCs work overtime,' you can read that 'Gateway's network of 8,000 PCs can deliver 14 teraflops.' This is plain wrong. You all know that this number of 14 teraflops is meaningless. It's just the addition of the peak speed of all the PCs -- never reached anyway on individual PCs. You need specialized software to work efficiently with a grid. And two companies are releasing new products to power grids. Avaki rolled out what it believes is the first Java-based data grid software for enterprise-class IT environments. Kontiki, for its part, on Monday released a grid server that brings its content delivery system into the server realm, whereas previously it was only available for PCs. Check this column for a summary, or this article for more details."
Why aren't those things called 'stick-up' ads, anyhow? Internet Ninja writes "Netscape today released version 7.01 of Netscape based on Mozilla 1.0.2. Back in is popup blocking which they got a lashing for in 7.0 as well as tabs as home pages just like Mozilla. Release notes here and there's a couple articles on Netscape devedge which may be of interest to developers."
And they will continue to have produced my Athlon, too. schnoz writes "And you thought AMD was quitting the PC chip market? Then check out this article on Business Week. Not only are they releasing new chips and plan to continue to do so, they're also still very active research wise, working on new transistor making techniques such as the double gate design as well as metal-rather-than-silicon design. Keep going at it AMD!!"
I was starting to worry when I heard that AMD wasn't going to compete anymore in the CPU market. They are going to have to try very hard to keep up with the new hyperthreading technologies from Intel, but who really needs a 3 GHZ CPU. Poor college students like me want something that is the most bang for the buck, and I haven't had any problems playing my games (battlefield 1942, UT2003) on my Athlon 1700 yet!!!!
Keep on chuggin' AMD. College students are behind you!
It's good to see (or bad to see) pop-up ads hitting the mainstream world.
Flamebait if I ever saw it..
The rambus technology is unique in that it was designed to co-operate with the P4 "netburst" architecture. That is, delivering very quick transfer of data, but the latency wasn't very good, and still isn't extremely good even though it is at 533mhz.
DDR is both cheaper and has better potential for the future. Do you ever see video cards (such as NV30 and ATI R300+) use rambus?
That and DDR2 is just around the corner. Basically, AMD has always followed the cheaper and faster route, and DDR is a lot cheaper than DDR.
My 2c.
Like, there's one where the mom is home alone with her little kid, and everyone knows that women are only motivated to actually cook when there's a hunky man around. So she's about to make the kid a FROZEN PIZZA when the kid holds up a drawing from school and says "Look, Mommy, I drawded you a pitcher!" and Mom oohs over it and to reward the kid she puts away the frozen pizza and instead the kid gets A BOWL OF CAMPBELL'S SOLID PINK "TOMATO" SOUP for lunch. This is love in the same sense that this is nutrition. Lumpless flesh-colored soup. Remember how Campbell's tried to use the slogan "Soup Is Good Food" for a few months until enough dieticians complained that that was an outright lie in the case of Campbell's watery slime? Remember how they got busted for always showing pictures of soup with the few measly pathetic little veggie bits standing on the surface of the soup because the bowls were always filled with GLASS MARBLES to hold up the little fragments of orange-gray carrots and caved-in peas?
Physicists are more than thinking about the Grid, I should know as they're funding my PhD in Data Grid Computing 8*).
The main reason for this is the Large Hadron Collider, which is due to go into production at CERN in about 2007. For the younger members of the audience, CERN was where Tim Berners-Lee developed the World Wide Web in the early 1990's
When it goes online it has 4 major experiments, each of which stores data at 100-400MB/sec, and I stress stores data at 100+ MB/sec, the first level is processing 40Terabytes a second. This equals a few petabytes a year (1PB = 1000TB = 1000000GB) which then has to be shipped to sites around Europe and the US.
All this is going to have data, processing and network requirements which make most techies gasp, i.e. Google only has a 20TB database, current physics ones are at 650TB+. At this level 14TFlops is kinda a cute little toy.
And yes, most of it's open source and based on the Globus Toolkit.
Note that it was just announced that Netscape is laying off people:
a ol _layoffs/
http://money.cnn.com/2002/12/10/news/companies/
Was this release of 7.01 just for spin, to try and keep the positve in the news more than the negative?
I hate marketing.
Using your sig line to advertise for friends is lame.
Now that Netscape has re-introduced popup blocking, Microsoft may soon follow suit. However, I did see an article on /. a while back about a group of advertisers that claimed any kind of blockage on their advertisements was theft (they claimed being able to see a site without having to see the ads constituted theft of bandwidth). If all future browsers incorporate popup blocking, where is the future of online advertising headed?
Rather than having window.open() return a null handle, have it return a real handle, but simply don't create the window. Better yet, have it optionally load the contents of the window, so the remote site never even knows that the window simply was never popped up.
Agreed. The physicists are putting massive resources into grid computing... trying to get all the scientific equipment onto a common paradigm of network resource access. I should also know, as I was co-opted into testing grid resources and grid algorithms when I was working for the University.
I think that there are a few more reasons than just the Hadron Collider, however.
The astrophysicists have got their Hubble Telescope and the radio telescopes which the SETI@Home project gets it's data from. The nuclear medical technicians have their magnetic resonance imagers (nMRI) and their picture archiving and communication systems (PACS). The geologists have got their seismographs. And the geneticists have got their DNA databases.
Surprisingly, a lot of scientific equipment is actually able to generate between 100MB to 1GB of data per second. Not just a collider or accelerator, although they are certainly known for generating alot of information.
Information is cheap and free, if you understand how to generate digital content. MRI scanners, for instance, are able to generate that much information, and are nearly always underclocked because physicians generally aren't looking at the atomic or molecular level.
Agreed on the Google point. It goes to show that high end computing is still order's of magnitude faster than home appliances (PCs). I was impressed at college with the virtual reality workstations we used to navigate the grid network (Internet2 connection, via the CERN group, Argonne National Laboratories, Enrico Fermi Institute, et al. I happend to have studied under one of the Globus Toolkit authors, working out of Argonne, for a very short while.).
Anyhow, a moderate scientific/medical workstation nowdays has perhaps 4 to 32 GB of RAM. We would use that RAM to generate immersive 3D rendering of nucleic acids, genomes, proteins, biotech designs, astrometic simulations, and so forth. Now, considering a stereoscopic projection monitor, you've got anywhere between 2 and 16 GB of data streaming to you, per second, per eye, via the projection monitor. Stereoscopic photorealistic virtual environments, grid overlays, you name it. It can definately be information overload at times.
The interesting thing, I found, was that at 2GB of information, per second, per eye, was the threshhold before I really and truly began to get mentally 'tricked' into being totally immersed, visually. That is, 20/20 vision, photorealistic, full spectrum color, stereoscopic, requires about 4 GB of memory to acurately and artificially calculate and project a pixel for each rod/cone in the human retina. A little bit of neurobiology for you, I suppose. Yep, them darn Turring machines are pretty neat, when they're hooked up to a grid network.
I went to a presentation yesterday by Platform (the guys that make LSF) who talked about grid computing. Each person that spoke gave a different definition of what 'grid computing' is: It's clusters of clusters, it's clusters plus processing on individual machines, etc.
The upside is that such processing using PCs is already taking place, in the form of distributed.net, folding@home and seti@home among many others. If gateway wants to use its spare cycles to create a supercomputer capable of many teraflops, then go for it.
On the other hand, apps that are well suited to such distributed computing are those that require little I/O and more number crunching. That is, you don't want to use BLAST (comparing gene sequences) as the data sets are on the order of GB. But simple number crunching, like the examples already given, do not require sending much data to the clients for processing.
BTW, LSF has software to do the same thing with desktop boxes.
As far as new ad formats, right on devedge page linked from the artice, you are seeing the future of web advertising.
Instead of popup windows (which are *SO* 90's), we will have popup div layers, positioned to cover the page. Look at Netscape's own popup detection example. They show you how to detect a popup blocker, and open up a fixed position DIV to give visitor a "warning". How long do you think it will take an ad network programmer to figure out that instead of the warning, this DIV can actually be used to show the ad itself?
Better yet, if the window failed to open, you can open the div with an IFRAME in it that points to the same URL. And no popups. :)
Welcome to the future. Doesn't it look a lot like the past?
Are you on drug(s)?!! Why not?
If it becomes plainly known that browser XYZ routinely "fakes" popups, people will figure out to use XYZ's strengths in order to still show the ads in another way.
Here's a test page I threw up yesterday as a proof of concept: LINK See how your Moz deals with that...
P.S. I thought I was pretty clever coming up with this, but a search revealed others already using it in live systems.