RAMCloud: The Idea of Storing All Data in RAM

[neiroatopelcc]

[citation][nom]ltdementhial[/nom]i really don't see an use for this type of things...although it will be nice if you are a Minecraft maniac...but in real life task even more than 8gb for now is very unnecessary and stupidly excessive.[/citation]

Minecraft server runs just fine on a reasonable magnetic disk, you don't need ram drives for that. Just get craftbukkit and a plugin to handle block writes. My server's capable of at least supporting 10 concurrent users without lag, and it's running off of 'old' phenom 2 tech with 7GB memory allocated to the game.

Oh and more than 8GB is NESSECARY for some applications. You try hosting an esx solution or a terminal server system with less. We used opterons with 9GB memory back in 2006 or so. Now the standard Xeon system carries 32GB with room to spare. Hell even my laptop needs at least 8GB to function properly (and needs a bigger ssd, but that's another story)

[citation][nom]amk-aka-Phantom[/nom]No really? This can be done for a consumer, as well. Imagine having your entire OS in the RAM? An X79 board has what, 8 RAM slots? Throw 8GB into each one and you get 64GB, that's more than enough for a full Windows installation and some programs. It will be blazing fast.[/citation]

Get a server board. Some of the nehalem based server boards are atx (most are ssi), and still support up to 32 memory slots. That's 256GB of memory. And in any event, almost all of them support at least 12 slots which is 96GB max.

 

[amk-aka-Phantom]

Get a server board. Some of the nehalem based server boards are atx (most are ssi), and still support up to 32 memory slots. That's 256GB of memory. And in any event, almost all of them support at least 12 slots which is 96GB max.

I was trying to stay within a more or less reasonable price limit 😀 But yes, of course you could do that. I honestly didn't know there're boards with 32 memory slots (how big must that thing be? Could you post a few links?), but since there are, surely you could make an even bigger RAM disk there...

 

[willwayne]

Static RAM, anyone?

 

[ojas]

[citation][nom]Krnt[/nom]Hmm... I tried to notice the stuttering in some heavier pages (newegg, modDB, devianart, TechpowerUP) with no success, then I returned to TH and now I notice it, maybe because now I opened a newer article that wasn't in the cache, I'm not sure why I didn't notice it before if it was present (maybe because I am a little accustomed to some slower machines lately), it happens just before showing the comments section (maybe the comments are loaded while scrolling).Also about the processor clock thing I was talking before, It idles at 800 MHz, jumps to 2.2 then to 2.7 and then to 3.7 depending on the usage, and for scrolling it just need to jump to 2.7 with a usage of 20% in that current state, about that, maybe I was't able to notice the stuttering because the CPU needs like 8 seconds to return to 800 from 3.7 GHz and if I make the test when the CPU is not in its lowest C-state the issue will not be noticeable, and maybe that is why in some Sandy CPUs it will not be noticeable.[/citation]

I think you're right, i checked, the stutter seems to be because SpeedStep kicks in and increases about 300-600 MHz, then falls back. Probably with higher clocks and a higher IPC it doesn't need to (or can't, like in your case when you disable it).
Either way can't put my finger on the reason other than the code on this site. I doubt the comments refresh on their own, afaik you have to reload the page for that. Now i'm no web designer, so i really can't say much...

[citation][nom]livebriand[/nom]I'm not on that machine right now, but on my AMD E350 netbook (also 4GB RAM@1066MHz, 320GB HD) with Windows 7 SP1 64-bit and Firefox 8, I tried scrolling up and down really fast and didn't notice anything. Actually, wait, there is a slight delay when I go by the top of the comments section. I have noticed that a little sometimes when scrolling through some webpages on here. But on my i5 machine, I don't notice it at all.[/citation]

Yeah seems to affect chrome more, what you describe on ff is what i experience on ie...that i5 thing...probably what Krnt says...IPC and such...maybe SB procs don't feel the urge to leave idle state??

Either way, i see coding issues...

 

[livebriand]

@ojas
I'm on that machine right now and using Chrome, and I can't notice that at all. Oh btw, this is NOT a sandy bridge, it's a 1st gen Lynnfield i5 750. It's staying at it's 1.2GHz idle speed right now.

 

[eddieroolz]

Basically, the whole thing is a huge RAM Disk then?

 

[spookie]

this is a very old idea! My dad has told me about something like this almost a year ago, SAP AG is busy experimenting with it. So sorry Stanford, you should try to be more inovative