Personal computing discussed

Moderators: renee, Flying Fox, Thresher

 
TheJack
Gerbil
Topic Author
Posts: 71
Joined: Sat Jul 12, 2014 8:06 pm

Does it make sense to make RAM RAIDable?

Sat Jul 12, 2014 8:17 pm

Or are there any major problems doing so? Any insights on this?
 
UberGerbil
Grand Admiral Gerbil
Posts: 10368
Joined: Thu Jun 19, 2003 3:11 pm

Re: Does it make sense to make RAM RAIDable?

Sat Jul 12, 2014 8:33 pm

Depending on what you mean by RAID, RAM already is or doesn't need it. If you mean operating in parallel like RAID-0, that's what dual-channel memory is doing (and has been for what, a decade on x86?). At the higher end, workstation and server chipsets have supported triple- or quad-channel memory; you don't see this at the low end because the performance doesn't justify the expense, particularly for consumer workloads where memory bandwidth just doesn't matter vs all the other things that can gate performance.

If you mean the traditional definition of RAID (the R stands for Redundant), various server architectures have redundant and/or hot-swappable RAM, and all of them support ECC.
 
TheJack
Gerbil
Topic Author
Posts: 71
Joined: Sat Jul 12, 2014 8:06 pm

Re: Does it make sense to make RAM RAIDable?

Sat Jul 12, 2014 9:16 pm

UberGerbil wrote:
Depending on what you mean by RAID, RAM already is or doesn't need it. If you mean operating in parallel like RAID-0, that's what dual-channel memory is doing (and has been for what, a decade on x86?). At the higher end, workstation and server chipsets have supported triple- or quad-channel memory; you don't see this at the low end because the performance doesn't justify the expense, particularly for consumer workloads where memory bandwidth just doesn't matter vs all the other things that can gate performance.

If you mean the traditional definition of RAID (the R stands for Redundant), various server architectures have redundant and/or hot-swappable RAM, and all of them support ECC.

Hi,
I thought dual channel is for increased bandwidth, which is not the same as RAID0. Hier is why I asked: with DDR4, stacking RAM will be possible. But I think heat could become a problem. If you could lower the voltage/frequency you may be able to overcome heat issues ( to some extend at least ) and compensate the decreased speed by RAID0ing the RAM layers. ????
 
deepblueq
Gerbil
Posts: 40
Joined: Sat May 17, 2014 9:32 pm
Location: the middle of nowhere

Re: Does it make sense to make RAM RAIDable?

Sat Jul 12, 2014 11:37 pm

How about this angle: RAID 0-ing two drives effectively doubles the width of the comm channel (the channel often isn't the limitation, but it looks the same from a performance perspective). The performance improvement is because when you would have been writing one bit, you're now writing two at once.

A stick of DDR3 RAM operates on 64 bits at a time, but by using two of them dual-channel, the processor is operating on data 128 bits at a time. This is almost exactly the same arrangement as with RAID 0.

Video memory interfaces are where you'll see this taken to extremes, with interfaces as wide as 512 bits.

We can get a ton of performance when parallelism is available (like with stacked chips), but at some point it takes more circuitry to handle it than it's worth. See video card memory interfaces - the amount of data a card can move around is every bit as major a limitation as how much number crunching it can do, but the manufacturers don't add more bandwidth, because that would use a lot of power and die space.
Pentium G3258 @ 4.3 | MSI GTX 650 Ti Boost | 8GB G.Skill DDR3-2133 | ASRock Z97E-ITX/ac | Crucial MX100 256GB | Corsair Graphite 380T
 
Flying Fox
Gerbil God
Posts: 25690
Joined: Mon May 24, 2004 2:19 am
Contact:

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 12:13 am

One effect of multi-channel memory vs chip stacking (Nvidia's vision, or the latest hypercube memory announcements) is indeed increased speed of data transfer. For multi-channel memory, you simply get more bits jammed through per unit of time. However, stacking also reduces footprint, shorten interconnects, and may even have thermal advantages.
The Model M is not for the faint of heart. You either like them or hate them.

Gerbils unite! Fold for UnitedGerbilNation, team 2630.
 
TheJack
Gerbil
Topic Author
Posts: 71
Joined: Sat Jul 12, 2014 8:06 pm

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 6:17 am

Thanks guys,
for what I know dual channel is not the same as raid0. Running RAM in dual channel gives a performance boost of around 5% whereas with raid0 you get almost twice the performance. raid0 is what SSDs are doing to achieve their speeds. in raid0 the controller divides the data to the number of available drives ( say, 5 ) and lets each drive do a fifth of the necessary work. so this is one thing I know. what is ambiguous ( to me ) is when it is compared to dual channel.
 
mattshwink
Gerbil Team Leader
Posts: 200
Joined: Wed Jul 16, 2008 7:54 am
Location: Alexandria, VA

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 6:45 am

There are some server boards that can mirror memory. It is only done for fault tolerance, as far as I know, and not for performance.
 
Kougar
Minister of Gerbil Affairs
Posts: 2306
Joined: Tue Dec 02, 2008 2:12 am
Location: Texas

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 7:20 am

TheJack wrote:
Thanks guys,
for what I know dual channel is not the same as raid0. Running RAM in dual channel gives a performance boost of around 5% whereas with raid0 you get almost twice the performance. raid0 is what SSDs are doing to achieve their speeds. in raid0 the controller divides the data to the number of available drives ( say, 5 ) and lets each drive do a fifth of the necessary work. so this is one thing I know. what is ambiguous ( to me ) is when it is compared to dual channel.


You're trying to compare the wrong metrics is the problem. Start by comparing the memory bandwidth, not real-world performance. For example take the same DDR3-1600Mhz RAM. For dual channel memory bandwidth it'd be ~21GB/s but for quad-channel memory, it jumps to ~37GB/s of available memory bandwidth. So yes it is like RAID 0 in that respect given it's an 80% increase in bandwidth.

The problem is real world performance doesn't increase by 80% to match the raw bandwidth because other parts of the system are the limiting factors, namely the processor. When reading from a disk, all it needs to do is transfer a massive amount of data to show a HD resolution photo onto the screen, minimal processing is required to do this so it is just mass-moving data. But when a processor is reading from RAM, it's usually doing so to actually perform computational work on whatever data it's moving. The CPU spends more time doing computational work than polling to the system RAM, and software design intentionally is optimized to minimize calls to system RAM in order to maximize the program's performance. So conceptually "RAIDing" memory as you put it won't remove these limiting factors. It's an apples-to-oranges comparison with what RAM versus disks do, even though both handle massive amounts of data. Now if you take a server or enterprise workload that constantly pushes gigabytes of traffic over the memory bus then quad channel would start to show appreciable real-world performance benefits. But you won't find that on most consumer desktops.

SSDs don't actually "RAID 0" the NAND flash either, if they did it would make them dangerous to use for storing data. And while I think all modern SSDs today use a form of RAID parity for error checking it still isn't done for performance.

Come to think of it, Intel's FB-DIMM was sort of like RAIDing RAM... the memory controllers talked to the Advanced Memory Buffer which sat between the CPU and RAM separating them, and acted like a RAID controller in that it converted the signal to/from serial as well enabled the ability to add far more addressable memory modules. The problem with that is it was very power in-efficient (any master controller will need to be powerful to handle DDR3 or DDR4 speeds with minimal latency penalty, and hence burn power) and it increased the total latencies involved (which RAID can also do). Not to mention it added to the system and special FB-DIMM costs... even that was done to just add memory capacity though, and as higher capacity DRAM was developed it became unneeded.
 
TheJack
Gerbil
Topic Author
Posts: 71
Joined: Sat Jul 12, 2014 8:06 pm

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 7:48 am

Kougar wrote:
TheJack wrote:
Thanks guys,
for what I know dual channel is not the same as raid0. Running RAM in dual channel gives a performance boost of around 5% whereas with raid0 you get almost twice the performance. raid0 is what SSDs are doing to achieve their speeds. in raid0 the controller divides the data to the number of available drives ( say, 5 ) and lets each drive do a fifth of the necessary work. so this is one thing I know. what is ambiguous ( to me ) is when it is compared to dual channel.


You're trying to compare the wrong metrics is the problem. Start by comparing the memory bandwidth, not real-world performance. For example take the same DDR3-1600Mhz RAM. For dual channel memory bandwidth it'd be ~21GB/s but for quad-channel memory, it jumps to ~37GB/s of available memory bandwidth. So yes it is like RAID 0 in that respect given it's an 80% increase in bandwidth.

The problem is real world performance doesn't increase by 80% to match the raw bandwidth because other parts of the system are the limiting factors, namely the processor. When reading from a disk, all it needs to do is transfer a massive amount of data to show a HD resolution photo onto the screen, minimal processing is required to do this so it is just mass-moving data. But when a processor is reading from RAM, it's usually doing so to actually perform computational work on whatever data it's moving. The CPU spends more time doing computational work than polling to the system RAM, and software design intentionally is optimized to minimize calls to system RAM in order to maximize the program's performance. So conceptually "RAIDing" memory as you put it won't remove these limiting factors. It's an apples-to-oranges comparison with what RAM versus disks do, even though both handle massive amounts of data. Now if you take a server or enterprise workload that constantly pushes gigabytes of traffic over the memory bus then quad channel would start to show appreciable real-world performance benefits. But you won't find that on most consumer desktops.

SSDs don't actually "RAID 0" the NAND flash either, if they did it would make them dangerous to use for storing data. And while I think all modern SSDs today use a form of RAID parity for error checking it still isn't done for performance.

Come to think of it, Intel's FB-DIMM was sort of like RAIDing RAM... the memory controllers talked to the Advanced Memory Buffer which sat between the CPU and RAM separating them, and acted like a RAID controller in that it converted the signal to/from serial as well enabled the ability to add far more addressable memory modules. The problem with that is it was very power in-efficient (any master controller will need to be powerful to handle DDR3 or DDR4 speeds with minimal latency penalty, and hence burn power) and it increased the total latencies involved (which RAID can also do). Not to mention it added to the system and special FB-DIMM costs... even that was done to just add memory capacity though, and as higher capacity DRAM was developed it became unneeded.


Nicely explained. I knew there was something wrong with my theory otherwise it would have already been done. So how on earth are they going to handle heat issues with stacked RAM?
 
Flying Fox
Gerbil God
Posts: 25690
Joined: Mon May 24, 2004 2:19 am
Contact:

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 8:02 am

Chip stacking is supposed to reduce power consumption. It seems like for RAMs it is more easily do-able.
http://www.extremetech.com/computing/15 ... -ddr-sdram
http://www.extremetech.com/computing/11 ... p-stacking

Our recent discussion of HMC: viewtopic.php?f=2&t=94402
The Model M is not for the faint of heart. You either like them or hate them.

Gerbils unite! Fold for UnitedGerbilNation, team 2630.
 
TheJack
Gerbil
Topic Author
Posts: 71
Joined: Sat Jul 12, 2014 8:06 pm

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 8:33 am

Flying Fox wrote:
Chip stacking is supposed to reduce power consumption. It seems like for RAMs it is more easily do-able.
http://www.extremetech.com/computing/15 ... -ddr-sdram
http://www.extremetech.com/computing/11 ... p-stacking

Our recent discussion of HMC: viewtopic.php?f=2&t=94402


Thanks. Good read.
 
just brew it!
Administrator
Posts: 54500
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 10:23 am

TheJack wrote:
I thought dual channel is for increased bandwidth, which is not the same as RAID0.

I think you are misunderstanding what RAID-0 is for. RAID-0 *is* done for increased bandwidth. It does not improve reliability; in fact it makes it worse, since the failure of a single drive causes you to lose the contents of both drives. RAID-0 is essentially a "dual channel" hard drive.

TheJack wrote:
Hier is why I asked: with DDR4, stacking RAM will be possible. But I think heat could become a problem. If you could lower the voltage/frequency you may be able to overcome heat issues ( to some extend at least ) and compensate the decreased speed by RAID0ing the RAM layers. ????

Latency matters too. As you decrease the clock speed, latency goes up, and adding more channels can't compensate for that.
Nostalgia isn't what it used to be.
 
TheJack
Gerbil
Topic Author
Posts: 71
Joined: Sat Jul 12, 2014 8:06 pm

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 10:38 am

just brew it! wrote:
TheJack wrote:
I thought dual channel is for increased bandwidth, which is not the same as RAID0.

I think you are misunderstanding what RAID-0 is for. RAID-0 *is* done for increased bandwidth. It does not improve reliability; in fact it makes it worse, since the failure of a single drive causes you to lose the contents of both drives. RAID-0 is essentially a "dual channel" hard drive.

TheJack wrote:
Hier is why I asked: with DDR4, stacking RAM will be possible. But I think heat could become a problem. If you could lower the voltage/frequency you may be able to overcome heat issues ( to some extend at least ) and compensate the decreased speed by RAID0ing the RAM layers. ????

Latency matters too. As you decrease the clock speed, latency goes up, and adding more channels can't compensate for that.



Right.
Nano holes may help with heat dissipation. Termites have a sophisticated system of ventilation where hot air is removed from their nests automatically through the interaction of hot and cold air. ( that would be too difficult though )
 
just brew it!
Administrator
Posts: 54500
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 10:55 am

TheJack wrote:
Nano holes may help with heat dissipation. Termites have a sophisticated system of ventilation where hot air is removed from their nests automatically through the interaction of hot and cold air. ( that would be too difficult though )

Nano holes is not much different than adding a heat sink. Either way, you are increasing the surface area through which heat can be dissipated by convection. The problem with nano holes is that they will quickly become clogged with dust.
Nostalgia isn't what it used to be.
 
TheJack
Gerbil
Topic Author
Posts: 71
Joined: Sat Jul 12, 2014 8:06 pm

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 11:15 am

just brew it! wrote:
TheJack wrote:
Nano holes may help with heat dissipation. Termites have a sophisticated system of ventilation where hot air is removed from their nests automatically through the interaction of hot and cold air. ( that would be too difficult though )

Nano holes is not much different than adding a heat sink. Either way, you are increasing the surface area through which heat can be dissipated by convection. The problem with nano holes is that they will quickly become clogged with dust.


The problem with a heat sink would be that it only works on the surface, not the heat accumulated between the layers. The increased surface area if done right could be kept at minimum. then we have the dust clogging issue. I think this could be taken care of by some sophisticated means. So basically I am getting to like the idea which I just brew. :wink:
 
Flying Fox
Gerbil God
Posts: 25690
Joined: Mon May 24, 2004 2:19 am
Contact:

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 11:38 am

Nanoholes with air cooling is still not going to cut it, as chips that so confined in a tiny area you will still have the problem of how to generate air flow to convect the heat out. Outfits like IBM have been researching nanoscale pipes to carry liquid for cooling chips. That hopefully gets us somewhere.

Sorry to burst your bubble, but I am sure there are many crazy scientists with similar (and more) ideas already. :P :lol:
The Model M is not for the faint of heart. You either like them or hate them.

Gerbils unite! Fold for UnitedGerbilNation, team 2630.
 
just brew it!
Administrator
Posts: 54500
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 11:49 am

Yeah, the nano holes would make more sense with a closed loop cooling system. That way you avoid the clogging issue, and can provide some additional means of moving coolant through the system... I bet a piezoelectric pump could work for this. Or skip the pump altogether, and make the chips themselves the hot end of a passive heatpipe (phase change) cooling loop.
Nostalgia isn't what it used to be.
 
TheJack
Gerbil
Topic Author
Posts: 71
Joined: Sat Jul 12, 2014 8:06 pm

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 11:57 am

Flying Fox wrote:
Nanoholes with air cooling is still not going to cut it, as chips that so confined in a tiny area you will still have the problem of how to generate air flow to convect the heat out. Outfits like IBM have been researching nanoscale pipes to carry liquid for cooling chips. That hopefully gets us somewhere.

Sorry to burst your bubble, but I am sure there are many crazy scientists with similar (and more) ideas already. :P :lol:


:) That was good.
But my bubble has still some air ( it is just air anyway ) . The problem with liquid cooling is that you need a means of circulating that liquid, whereas with air ventilation it can be done through the simple interaction of hot and cold air, just like termites do it. Maybe that is why IBM has had no luck with their approach.
 
just brew it!
Administrator
Posts: 54500
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 12:10 pm

You're not going to get enough natural air convection in a tightly confined space (your nano holes) for it to make much of a difference. And as previously noted, the holes will quickly get plugged by dust.

If you don't want a pump, use a heat pipe; they circulate their coolant without needing a mechanical pump. The temperature gradient itself essentially "pumps" the coolant inside the heat pipe by causing a phase change from liquid to gas.
Nostalgia isn't what it used to be.
 
TheJack
Gerbil
Topic Author
Posts: 71
Joined: Sat Jul 12, 2014 8:06 pm

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 12:20 pm

just brew it! wrote:
You're not going to get enough natural air convection in a tightly confined space (your nano holes) for it to make much of a difference. And as previously noted, the holes will quickly get plugged by dust.

If you don't want a pump, use a heat pipe; they circulate their coolant without needing a mechanical pump. The temperature gradient itself essentially "pumps" the coolant inside the heat pipe by causing a phase change from liquid to gas.


Anyway, I think with chip stacking, cooling is going to get more and more attention. Let's see how they will get around that.
 
LASR
Gerbil First Class
Posts: 147
Joined: Fri Jan 10, 2014 9:35 pm

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 1:12 pm

They already use chip stacking in mobile SOCs. In this case, they are actually stacking the CPU under the RAM. Cooling is not a problem unless you are working with very small thermal envelopes such as in tiny cell phones. Even on slightly larger tablet devices, passive cooling is enough to keep these devices from throttling.

Since RAM has a much lower TDP than SOCs, I am going to assume multi layer chip stacking will not be a problem with desktop-class cooling solutions.

Having said that, I believe you have some misconceptions about the notions of what RAID and multi channel memory actually mean. I would suggest reading up about what they are first, before trying to "make RAM RAIDable."
 
TheJack
Gerbil
Topic Author
Posts: 71
Joined: Sat Jul 12, 2014 8:06 pm

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 1:29 pm

LASR wrote:
They already use chip stacking in mobile SOCs. In this case, they are actually stacking the CPU under the RAM. Cooling is not a problem unless you are working with very small thermal envelopes such as in tiny cell phones. Even on slightly larger tablet devices, passive cooling is enough to keep these devices from throttling.

Since RAM has a much lower TDP than SOCs, I am going to assume multi layer chip stacking will not be a problem with desktop-class cooling solutions.

Having said that, I believe you have some misconceptions about the notions of what RAID and multi channel memory actually mean. I would suggest reading up about what they are first, before trying to "make RAM RAIDable."


The question with ram and raid0 has been answered. My next question is about the newly introduced Samsung stacked SSDs. 32 laqyer, but how thick is each layer, has anyone torn one apart to see what it looks like?
 
just brew it!
Administrator
Posts: 54500
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 1:35 pm

Pretty sure the Samsung 3D flash stack is all inside the flash chip package, not multiple separate packages stacked and soldered like with the SOC/RAM used in cell phones. You'd probably need to X-ray the chips to see anything interesting.
Nostalgia isn't what it used to be.
 
Crayon Shin Chan
Minister of Gerbil Affairs
Posts: 2313
Joined: Fri Sep 06, 2002 11:14 am
Location: Malaysia
Contact:

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 1:49 pm

I think in the context of RAM, RAID0 striping is called bank interleaving.

Dual channel is basically taking two existing sets of 64 wires, and treating them as one large pipe instead of two separate pipes. Sounds similar to RAID0, but has nothing to do with splitting up data and all to do with having an effectively wider path for data to flow across. Whereas RAID0 achieves that by striping.
Mothership: FX-8350, 12GB DDR3, M5A99X EVO, MSI GTX 1070 Sea Hawk, Crucial MX500 500GB
Supply ship: [email protected], 12GB DDR3, M4A88TD-V EVO/USB3
Corsair: Thinkpad X230
 
TheJack
Gerbil
Topic Author
Posts: 71
Joined: Sat Jul 12, 2014 8:06 pm

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 1:58 pm

just brew it! wrote:
Pretty sure the Samsung 3D flash stack is all inside the flash chip package, not multiple separate packages stacked and soldered like with the SOC/RAM used in cell phones. You'd probably need to X-ray the chips to see anything interesting.


In other words, in terms of thickness, you couldn't tell the difference between a normal and stacked SSD?
 
just brew it!
Administrator
Posts: 54500
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 2:14 pm

TheJack wrote:
just brew it! wrote:
Pretty sure the Samsung 3D flash stack is all inside the flash chip package, not multiple separate packages stacked and soldered like with the SOC/RAM used in cell phones. You'd probably need to X-ray the chips to see anything interesting.

In other words, in terms of thickness, you couldn't tell the difference between a normal and stacked SSD?

I have no idea whether you can tell the difference just by looking at the chips, but at most it is probably just a slightly taller chip package.
Nostalgia isn't what it used to be.
 
LASR
Gerbil First Class
Posts: 147
Joined: Fri Jan 10, 2014 9:35 pm

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 2:24 pm

TheJack wrote:
just brew it! wrote:
Pretty sure the Samsung 3D flash stack is all inside the flash chip package, not multiple separate packages stacked and soldered like with the SOC/RAM used in cell phones. You'd probably need to X-ray the chips to see anything interesting.


In other words, in terms of thickness, you couldn't tell the difference between a normal and stacked SSD?


They will look like any other chip. PoP (Package on Package) layering does indeed add to the height of the chips. But it will be very hard to tell the difference unless you have precision instrument. But I suspect Samsung are not stacking the chips using PoP. There are a variety of stacking techniques. I don't think they've disclosed what they use.
 
Flying Fox
Gerbil God
Posts: 25690
Joined: Mon May 24, 2004 2:19 am
Contact:

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 3:50 pm

Crayon Shin Chan wrote:
I think in the context of RAM, RAID0 striping is called bank interleaving.

Dual channel is basically taking two existing sets of 64 wires, and treating them as one large pipe instead of two separate pipes. Sounds similar to RAID0, but has nothing to do with splitting up data and all to do with having an effectively wider path for data to flow across. Whereas RAID0 achieves that by striping.

In dual channel configuration, the "bank interleaving" happens at the stick level afaiu.
The Model M is not for the faint of heart. You either like them or hate them.

Gerbils unite! Fold for UnitedGerbilNation, team 2630.
 
TheJack
Gerbil
Topic Author
Posts: 71
Joined: Sat Jul 12, 2014 8:06 pm

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 5:25 pm

Thanks guys. Awesome 8)
 
just brew it!
Administrator
Posts: 54500
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: Does it make sense to make RAM RAIDable?

Sun Jul 13, 2014 7:53 pm

Flying Fox wrote:
Crayon Shin Chan wrote:
I think in the context of RAM, RAID0 striping is called bank interleaving.

Dual channel is basically taking two existing sets of 64 wires, and treating them as one large pipe instead of two separate pipes. Sounds similar to RAID0, but has nothing to do with splitting up data and all to do with having an effectively wider path for data to flow across. Whereas RAID0 achieves that by striping.

In dual channel configuration, the "bank interleaving" happens at the stick level afaiu.

Bank interleaving is an entirely different thing, and has nothing to do with the number of memory channels on the motherboard.

Memory *chips* contain multiple banks of memory cells internally. Bank interleaving refers to the practice of reading/writing data from/to these banks in sequence, and overlapping the operations to increase throughput. While bank A is being accessed, bank B (the bank to be accessed next) is already being prepared for access.

Dual (or more) channel memory buses on the motherboard allow multiple sets of DIMMs to be accessed simultaneously. This causes effective memory bandwidth to scale with the number of channels. Multi-channel memory controllers can operate in two different modes -- ganged or unganged:

- In ganged mode, the system behaves as if it has a single memory bus, but with a data bus width equal to 64x the number of channels (a standard non-ECC DIMM is 64 data bits wide). The channels all operate in lockstep, but you can fetch more data (128, 192, or 256 bits, depending on whether the memory controller is 2, 3, or 4 channel) with each memory access.

- In unganged mode, each 64 bit channel operates independently. A channel can fetch data from anywhere in the DIMM(s) connected to that channel, and does not need to run in lockstep with the other channels (though in the case where you're streaming lots of data sequentially to/from DRAM it may still end up behaving much like ganged mode). The advantage of unganged mode is that it can be more efficient when there are lots of small, scattered memory accesses from multiple threads.
Nostalgia isn't what it used to be.

Who is online

Users browsing this forum: No registered users and 1 guest
GZIP: On