Something that I'm convinced of is that the processors are going to migrate to where the transducers are. Thus, every display will be intelligent; every NIC will be intelligent; and, of course, every disk will be intelligent. I got the "smart disk" religion from you, Dave. You argued that each disk will become intelligent. Today each disk has a 200-megahertz processor and a few megabytes of RAM storage. That's enough to boot most operating systems. Soon they will have an IP interface and will be running Web servers and databases and file systems. Gradually, all the processors will migrate to the transducers: displays, network interfaces, cameras, disks, and other devices. This will happen over the next decade. It is a radically different architecture.
What I mean by that is it's going to have a gigahertz or better processor in it. And it will have a lot of RAM. And they will be able to run almost any piece of software that you can think of today. It could run Oracle or Exchange or any other app you can think of.
In that world, all the stuff about interfaces of SCSI and IDE and so on disappears. It's IP. The interface is probably Simple Object Access Protocol (SOAP) or some derivative of SOAP; you send requests to it and get back responses in a pretty high-level protocol. The IP stack does security and naming and discovery. So each of these "disks" will be an IP version 6 (IPv6) node—or IPv9.
future of computing?
-
Jonathan
- Grand Pooh-Bah
- Posts: 6722
- Joined: Tue Sep 19, 2006 8:45 pm
- Location: Portland, OR
- Contact:
future of computing?
So here's an interesting viewpoint from some guy who won a Turing award:
-
quantus
- Tenth Dan Procrastinator
- Posts: 4891
- Joined: Fri Jul 18, 2003 3:09 am
- Location: San Jose, CA
He's probably right in everything except the time frame... I'd push it off for at least 15-20 years instead. Things do seem to be trending in that direction though. There's always been the trend of bringing the computation to the data, not the other way around. We've been trying to pull data to the computational units and keep losing the fight. We instead resort to moving even closer to memory. For example, caches used to be on separate chips, now they're on the same chip. Moving data is just too expensive compared to computing on it. This isn't true for everything yet, but for more and more applications, this is becoming true.
I think he mis-speaks a little by saying the processing will move to the network interfaces since it's going to be the network interfaces that will be doing the migrating with the processors. Everything will be able to do some reasonable and appropriate amount of processing and be able to communicate with the world at some reasonable and appropriate data-rate. Maybe he meant that processing will move to gateways, which is also already true in order to ensure QoS and even try to detect and stop DDOS attacks.
Everything having a GHz processor in 10 years is pretty likely, but it's not going to be enough to bring about the revolution he seems to be aluding to. BTW, I think it would be awesome to have smart traffic lights and stuff, especially if I could convince them to be green whenever I'm coming.
I think he mis-speaks a little by saying the processing will move to the network interfaces since it's going to be the network interfaces that will be doing the migrating with the processors. Everything will be able to do some reasonable and appropriate amount of processing and be able to communicate with the world at some reasonable and appropriate data-rate. Maybe he meant that processing will move to gateways, which is also already true in order to ensure QoS and even try to detect and stop DDOS attacks.
Everything having a GHz processor in 10 years is pretty likely, but it's not going to be enough to bring about the revolution he seems to be aluding to. BTW, I think it would be awesome to have smart traffic lights and stuff, especially if I could convince them to be green whenever I'm coming.