Monday, September 13th 2021

ZeroPoint Technologies wants to Compress the Data in your RAM

Some of you might be old enough to remember various "RAM doubling" software software solutions that appeared back in the late 1980's for Apple, as well as in the mid 90's for Windows 95 computers. Most of them never really delivered on their claims, but now it looks like we might be getting something similar, albeit in hardware.

Swedish company ZeroPoint Technology AB has announced that it has raised €2.5 million in a seed round to bring its Ziptilion patented memory compression technology IP to the market. ZeroPoint claims a compression ratio of two to three times depending on the workload, which seems very impressive. Unlike current software compression technologies such as ZSWAP or ZRAM that are used to compress data in RAM at a rate of 1.4 to 1.5 times, ZeroPoint promises that it's hardware IP won't have any real world negative effects on system performance. In fact, they claim it'll only cause one nanosecond of extra latency when writing data and 100 nanoseconds delay when it comes to reading the compressed data from RAM.
Ziptilion has already been implemented into a product using TSMC's 28 nm node, using an AXI bus at 800 MHz, which resulted in a memory throughput of 32 GB/s. It's not clear what product this is, but based on the manufacturing node it suggests it might just be an ARM based test platform. ZeroPoint claims that its IP is easy to integrate with existing hardware designs, as it's placed on the memory access path and thus doesn't require changes to the memory controller itself, although it does require a software device driver element which might limit some potential applications.

In something like a normal desktop computer on a 7 nm node, the Ziptilion IP is expected to only use up about 1.36 square millimetres for a standard dual-channel RAM configuration. On a server CPU with 8-channels of RAM, this is estimated to increase to about 3.02 square milimetres. ZeroPoint is expecting it's IP to find its way into servers, embedded systems and smartphone SoCs to start with and the company is already working on its next generation Ziptiolion+ technology which appears to further increase the compression factor depending on the specific target application. Not all kinds of data compress the same, based on ZeroPoint's whitepaper.
The boldest claim made by ZeroPoint is that they claim you'll see the same system performance gains in most of their tested scenarios, as you would from doubling your system RAM. In addition to this, ZeroPoint also claims to increase the effective memory bandwidth in a system, due to the compressed data taking up less bandwidth when being transferred between the RAM and the CPU. The whitepaper also seems to hint at the possibility of Ziptilion being used with GPUs, at least in something like an SoC, where the RAM is shared between the CPU and GPU.

It all sounds a little bit too good to be true, but the company has been researching its memory compression algorithms for the past 15 years, so it doesn't sound like a money grabbing scam, unlike the RAM doubling software back in the day.
Source: ZeroPoint Technology AB
Add your own comment

25 Comments on ZeroPoint Technologies wants to Compress the Data in your RAM

#2
lexluthermiester
Data compression in RAM? Why? Today, RAM is reasonably priced for most people and use-case scenario's. 20 years ago this would have been useful. Today? Dubious...
Posted on Reply
#3
Prima.Vera
lexluthermiesterData compression in RAM? Why? Today, RAM is reasonably priced for most people and use-case scenario's. 20 years ago this would have been useful. Today? Dubious...
This is a solution for servers mostly, not for average Joe. There RAM really matters. ;)
Posted on Reply
#4
Tardian
lexluthermiesterData compression in RAM? Why? Today, RAM is reasonably priced for most people and use-case scenario's. 20 years ago this would have been useful. Today? Dubious...
ZeroPoint is expecting it's IP to find its way into servers, embedded systems and smartphone SoCs to start with
So RAM limited applications?
Posted on Reply
#5
lexluthermiester
Prima.VeraThis is a solution for servers mostly, not for average Joe. There RAM really matters. ;)
While that's fair, how bad is it really? The servers I work with rarely reach full allocation. Granted those are servers with 2TB per rack, but still..
Posted on Reply
#6
londiste
It all sounds a little bit too good to be true, but the company has been researching its memory compression algorithms for the past 15 years, so it doesn't sound like a money grabbing scam, unlike the RAM doubling software back in the day.
If they found an algorithm that is useful in most cases and detrimental in relatively few cases and are able to implement this in hardware (which should make it possible to achieve the low latency addition they claim) it does not sound too good to be true.

Especially with them talking about adding their IP to the CPU side with memory controller - wouldn't it be bandwidth boost as well as size? GPUs have been using DCC for a while now for similar purpose. Although at a storage level, hardware compression of storage data before sending it over PCIe to NVMe SSD - that both new generation consoles do and DirectStorage will support on PC soon - falls into similar category.
Posted on Reply
#7
rutra80
If you store data in RAM you want max speed. If speed is not a priority, swap to SSD.
I don't remember RAM compression being a thing on Amiga...
Posted on Reply
#9
rutra80
@TheLostSwede It's in game/gag category which is for joke-programs. It probably just patches the system to show double the overall memory. I'd remove Amiga mention from the article ;-)
Posted on Reply
#10
TheLostSwede
rutra80@TheLostSwede It's in game/gag category which is for joke-programs. It probably just patches the system to show double the overall memory. I'd remove Amiga mention from the article ;-)
Fair enough.
Posted on Reply
#11
lexluthermiester
rutra80If speed is not a priority, swap to SSD.
Oh good grief no. That's not a good idea.. The less wear we inflict on our SSDs the better.
rutra80I don't remember RAM compression being a thing on Amiga...
It wasn't.
Posted on Reply
#12
windwhirl
lexluthermiesterData compression in RAM? Why? Today, RAM is reasonably priced for most people and use-case scenario's. 20 years ago this would have been useful. Today? Dubious...
Didn't stop Microsoft from adding it on Windows 10. Granted, it's only used for things that are just loaded in memory but not currently in use, but the feature is still there.
Posted on Reply
#13
xorbe
This likely won't mix with cloud servers with encrypted memory. Hard pass.
Posted on Reply
#14
trparky
Windows 10 already does some limited data compression in system RAM. For instance...
Posted on Reply
#15
rutra80
IIRC Windows compresses only what is paged - it doesn't compress working set.
Posted on Reply
#16
First Strike
trparkyWindows 10 already does some limited data compression in system RAM. For instance...
This is a different layer of technology. Windows compresses inactive memory regions. When it becomes active, Windows decompress it and place it back, uncompressed, into the memory, then resume operation. So it is a memory management strategy, a common OS capability, just more sophisticated. In older OS, they just swap the mess into HDD.

Whole memory encryption/compression is a hardware feature. Ideally it should be transparent to all upper layers and all computations. No matter inactive/active, OS/user level. It is a hardware architecture design.

A direct contrast: There is one region of memory that Windows 10 can never ever compress: itself.
Posted on Reply
#17
lexluthermiester
windwhirlDidn't stop Microsoft from adding it on Windows 10. Granted, it's only used for things that are just loaded in memory but not currently in use, but the feature is still there.
Right, but we're not talking about inactive data, the article above is talking about real-time full memory space compression. I'm not saying it wouldn't be useful in some situations, but unless such a scheme can be done by an ASIC on DIMM, there will be a system resource overhead which will eat into overall system performance. And I say no thank you. RAM is cheap. Just get more..
rutra80IIRC Windows compresses only what is paged - it doesn't compress working set.
If it's paged, it's not in RAM, that's the whole point of a pagefile/swapfile.
Posted on Reply
#18
rutra80
lexluthermiesterIf it's paged, it's not in RAM.
Nope. W10 has memory store which is a kind of compressed page file in RAM. When memory manager figures something out of the process' working set, it swaps it to that store in compressed form. If memory pressure increases it swaps it to the page file (still in compressed form).
Posted on Reply
#19
lexluthermiester
rutra80Nope. W10 has memory store which is a kind of compressed page file in RAM.
That would be a RAM store, not a page file.
rutra80When memory manager figures something out of the process' working set, it swaps it to that store in compressed form. If memory pressure increases it swaps it to the page file (still in compressed form).
Right, so how was the above statement incorrect?
Posted on Reply
#20
bonehead123
TheLostSwedecompany is already working on its next generation Ziptiolion+
Anyone know exactly how many "+" signs we can expect this so-called new tech to go through before we can actually say that it works or doesn't work ?

Just wondering if they are planning to follow intel's skylake/14nm ++++++++++++++++++ path & milk it for every penny they can get, hehehehehe :)
Posted on Reply
#21
rutra80
lexluthermiesterThat would be a RAM store, not a page file.


Right, so how was the above statement incorrect?
"1st stage" of paging is a RAM store.
Posted on Reply
#22
Mussels
Moderprator
wait til Qualcomm buys them out and your 16GB RAM phone turns out to have 8, compressed
Posted on Reply
#23
Jism
lexluthermiesterData compression in RAM? Why? Today, RAM is reasonably priced for most people and use-case scenario's. 20 years ago this would have been useful. Today? Dubious...

It was very common back in the days to use a util like this. Software for "only" 49.95$ or so instead of a 4MB additional ramstick for 200 to 400$.

But it would add extra latency, cycles etc and back in the days with such hardware it was a huge performance impact compared to todays standards.

Ram compression at hardware level could actually work. You could dense far more data into a stick then without. Imagine if this would work on GPU's too. You could save perhaps half of the bandwidth normally needed for graphics or compute operations. Not something small.
Posted on Reply
#24
lexluthermiester
Jism

It was very common back in the days to use a util like this. Software for "only" 49.95$ or so instead of a 4MB additional ramstick for 200 to 400$.
Did you actually WATCH that video? Yeah, you should actually watch it and then come back and think your statement over..
JismBut it would add extra latency, cycles etc and back in the days with such hardware it was a huge performance impact compared to todays standards.

Ram compression at hardware level could actually work. You could dense far more data into a stick then without. Imagine if this would work on GPU's too. You could save perhaps half of the bandwidth normally needed for graphics or compute operations. Not something small.
I understand how compression works and the context at play in the article. You're missing a few points though..
Posted on Reply
#25
Pepamami
lexluthermiesterData compression in RAM? Why? Today, RAM is reasonably priced for most people and use-case scenario's. 20 years ago this would have been useful. Today? Dubious...
Linux distros use compression for ram disks/devices, like temp folders or swap pages stored in ram (instead of disk) e.c zram/zswap, and reading compressed data from ram is faster, than read data from swap file placed on ssd.
Posted on Reply
Add your own comment
Copyright © 2004-2021 www.techpowerup.com. All rights reserved.
All trademarks used are properties of their respective owners.