Faulty Marvell Chips Delay SATA 6G Launch 90
Vigile writes "The SATA 6G standard offers more than simply a faster 6.0 Gb/s data throughput speed, to wit: improved NCQ support, better power management, and a new connector to support 1.8-inch drives. While modern-day, spindle-based hard drives struggle to keep up with SATA 3G speeds, modern SSDs are nearly saturating the existing standard, and a move to SATA 6G was welcome in the hardware community. It looks like that technology will be delayed, though. The only chip supporting the standard today, the Marvell 88SE9123, is having major issues. Motherboard vendors including ASUS and Gigabyte, which had planned on releasing SATA 6G technology using the chip on Intel Lynnfield platform motherboards later this summer, are having to remove the Marvell 88SE9123 and redesign their boards at the last minute due to significant speed and reliability issues."
Interface speed only (Score:1, Informative)
This is silly. Current hard drives can't even max out the speed of SATA-2. There's no need for faster interfaces just yet.
Re:Interface speed only (Score:4, Informative)
... "modern SSDs are nearly saturating the existing standard" ...
Re: (Score:3, Insightful)
*cough*Doesn't increase bandwidth*cough*
Re: (Score:1)
*cough* Benadryl ...
Re:Interface speed only (Score:5, Funny)
*cough*Regular mints are better if it's just a cough. No point in taking allergy medicine if you aren't allergic.
Re: (Score:1)
What you need is a Ricola.
Re: (Score:2)
This may not be the case anymore, but I had thought support for port expanders on consumer grade chipsets were pretty poor.
Re: (Score:2)
I have yet to see an SSD that can constantly saturate a 3.0Gbps link, please show me some that do.
Re: (Score:3, Interesting)
TFA said "nearly saturating", but regardless...
http://www.anandtech.com/storage/showdoc.aspx?i=3531&p=24 [anandtech.com]
250MB/s is just below the theoretical limit of SATA 3Gb/s, which is 300MB/s. It's possible that there are still other bottlenecks beside the hard drive.
I also have seen SSD RAID benchmarks somewhere, but I don't remember where.
Re:Interface speed only (Score:5, Informative)
Intel's X25-M [tomshardware.com] seems to perform 200 MB/s constant throughput. Granted that's "only" 2/3rds of the 3 Gbps that SATA 2 delievers, but the quote was "nearly saturate".
And if we're already at 2/3rds, that's a fairly compelling argument to upgrading. On laptops it can become an issue much quicker, as you usually only have 1 eSATA port, and port multipliers do not increase bandwidth. Hooking two X25-Ms onto a single eSATA 2 port can saturate it while doing non-random transfers easily but still have room left over on eSATA 3.
But if we're merely talking SSD in general, we can always point to Fusion-io ioDrive [tomshardware.com] which bottoms out at 429 MB/s.
Re:Interface speed only (Score:5, Informative)
Most sequential throughput benches hit the drive with sequential requests, and are not multithreaded. You end up seeing throughput lower than the theoretical interface bandwidth because of the latency involved for each request. A queue depth of 2 or 4 will give an X25 enough heads up to truly saturate the bus. I've recorded as high as 285 MB/sec from an X25-M using an NCQ-capable benchmarking tool.
Also, don't forget SATA uses 8/10b encoding, so you have to account for that overhead as well when calculating theoretical maximum throughput. Doing the math, you'll find 285 comes to 95% of a 3 GB/sec SATA interface. That's way higher than 200 MB/sec and close enough to call saturated.
Allyn Malventano
Storage Editor, PC Perspective
Re: (Score:3, Informative)
The Fusion-io ioDrive Duo [tomshardware.com] could, but connected directly into a PCI-E slot rather than SATA, but it is definitely possible to make a drive that will.
Re: (Score:2, Informative)
Can't boot from them, but there you go.
Re: (Score:2)
I looked at the Fusion-io cards a while ago, the speed is mind-blowing but the fact that it needs a driver made it an immediate no-go. Bootability is not the issue, it's all about compatibility. Why they can't make it appear as a plain old SATA or SCSI controller, I do not know, but more importantly I don't care, especially when you consider the asking price for these things. Even the cheapo Gigabyte I-Drive pretended to be a SATA disk (it literally had a SATA output connector). I'd be happy if this thi
Re:Interface speed only (Score:5, Interesting)
What manufacturer is going to make SATA SSD's that can saturate the fastest SATA port?
Sounds to me like a bad idea. Surely there are some tradeoffs between speed and some other also important metric. Ex, faster might mean larger erase block sizes.
It is likely that Intel could have made their product much faster, but without any benefit at all to doing so, they wouldn't.
This new SATA, while important, it sort of too-little-too-late. We need a much higher ceiling, and we need it yesterday.
Re: (Score:2)
"Sounds to me like a bad idea. Surely there are some tradeoffs between speed and some other also important metric. Ex, faster might mean larger erase block sizes."
But this has nearly no effect on reads, which is the majority of what most drives are doing. You want the fastest interface possible, we're talking about *flash memory* here.
There is no drawback to faster interfaces, your concerns are irrelevant since you're talking about engineering flaws and the immaturity of solid state, the interface is immat
Re: (Score:3, Insightful)
What manufacturer is going to make SATA SSD's that can saturate the fastest SATA port?
The same idiot manufacturers who deliver super-high sequential reads at the expense of everything else. There's a jumble of drives running off a particular J-Micron controller (or two :P), that deliver faster sequential reads than the Intel X25-E, but have random access times in the 50ms range (average). That's five times slower than a cheap 5400rpm notebook drive! They still sell like hotcakes because they're cheap, and we all know how much America loves cheap garbage. You could stick a SATA port out t
Re: (Score:2)
Better get used to it though, SATA is the new bottleneck, and it will be for years to come. There's just no way around it.
You mean aside from faster storage topologies which exist for people who need them.
Re: (Score:2)
As a hard drive interconnect ?
Sure, you can RAID a bunch of drives in an external chassis and pipe the combined throughput over FC, but how many people do you know with an FC HBA in their gaming rig or media workstation ? Even ignoring the minor detail that an entry-level HBA costs more than most people's entire PCs, you simply cannot compare external vs internal storage solutions. Some of us just don't want another big noisy box under our desks. It's fine for a datacenter but certainly not for SOHO.
Re: (Score:2)
SOHO users need more than 6Gbps?
Re: (Score:2)
Only 1, and he did it because he could keep the loud drive cabinet in the basement and use the drives from his bedroom. It was decent gear, too, LSI PCI-E 4GBps card with two multimode transceivers to a JBOD cabinet in the basement.
Re: (Score:2)
I suck, I meant 4 Gbps.
Re: (Score:2)
That's an awful lot of words.
Next time you could just say "SATA sucks".
Re: (Score:2)
I think its more likely that when SATA600 was on the drawing board, that nobody predicted this new market.
Stupidity is more likely than malice.
Re: (Score:2)
Sounds to me like a bad idea. Surely there are some tradeoffs between speed and some other also important metric. Ex, faster might mean larger erase block sizes.
Progress is not made by people who concoct imaginary dangers over yonder hill, and then use them as an excuse not to set out at all.
Re: (Score:2)
Why? You didn't take a long hard think about it at all.
If you can't saturate the link with one drive, and the standard procedure is to hook one drive to one port, then why do we need something faster now?
Something faster is fine, would even help burst speeds, but its not something we 'need' nor was it 'needed' yesterday for anything other than some silly setup that was never designed to be fast in the first place ... i.e. using a port replicator on a laptop.
Re: (Score:1)
If you can't saturate...
Cart before the horse. I can't saturate it because nobody is making drives that can.
I claim that it is no coincedence that SSD's, once performance became a comodity (thank you MTRON), quickly shot up from ~16MB/sec and leveled off at 200+MB/sec sequential read speeds. In this time there was no substantial technological breakthrough that made this possible, and we dont need one to go faster. What we need is a faster pipe.
As evidence of this fact, I will draw your attention to the future. Once SATA600 is
Re: (Score:3, Insightful)
Build it and they will come... (Score:2)
There is no way a manufacturer is going to produce a drive with Sata 2 specs that would exceed 300 Mb/sec. It would be a waste of money on their part (pearls before swine). Give a proper pipeline and room to breath and watch competition force the issue. ANY increase in pipeline is a good thing because the manufacturer's will move in to fill the pipe or lose out on the performance game.
Re: (Score:2)
"The only reason SSDs aren't capable of higher speeds is because the bus is not capable of more. There's no point making a controller capable of 2GB/s if you are only able to transfer at 300MB/s."
This is idiotic, every bus interface always has more bandwidth than most devices can take advantage of, the case was the same when we moved from ISA to PCI and then to PCIe (peripheral bus), from IDE to EIDE to ATA to SATA (hard disk bus).
Re: (Score:2)
Re: (Score:2)
Regular HDDs saturate SATA 3.0Gbps links (Score:3, Insightful)
Re: (Score:1)
This lack of knowledge on /. is sickening
That's because it's become a warren for sub-29 year olds who use bing.com as their SE of choice and actually brag about it :D.
[yawns/arches back] it's about time for my cuppa tea and afternoon nap.
Failure (Score:1)
Maybe they should stick... (Score:4, Funny)
Maybe they should stick to comic books.
Re: (Score:1)
I welcome less sophomoric word play.
Re:Maybe they should stick... (Score:5, Informative)
Um, Billy Batson would have you know it was Tesla and Edison, not Einstein, arguing over type of current.
Nice try.
Re: (Score:1)
Yes, but then we'd have more WHOOSH moments, you insensitive clod :P
Re: (Score:2)
Unlike like their ethernet chip, which absolutely will stop on a 64bit machine with over 4GB.
Re: (Score:2)
Noone's surprised. Marvell's had a track record of faulty, ill-documented, or bug-plagued parts. Sometimes, I wonder why they still bother. I suppose, someone has to make Realtek look good.
Re: (Score:2)
A couple years ago when Asus first started puting Marvell network chips on their boards, every single one that I saw (I or friends bought) had a faulty NIC. The symptom? It was slow...paaaaainfully slow...definitely not 1Gbps. More in the range of 10Kbps. I called up suppliers and Asus directly...they all said to RMA the motherboard, no DOA replacement. Joy. So, spend a stupid amount of time to remove the board, live witho
Don't Delay, Just Rename (Score:4, Funny)
Instead of delaying the launch, they could just rename the chip. SATA 5.5G for example.
Re: (Score:1)
Perhaps we could just call it 6gS and add video capabilities!?
Re: (Score:1)
Dammit.
I've got 12 2Tb SATA6 HD's here just waiting to be hooked up to a bus that can handle their speed!
Re:Don't Delay, Just Rename LOL!!!! (Score:1)
ROTF, LMMFAO.... Made my day, neonprimetime...
That's really kewl....especially in light of the recent story about companies manipulating or avoiding or playing games with certain release numbers.
Faulty Marvell Chips, eh? (Score:3, Funny)
I wonder if The Green Gobllin [marvel.com] had anything to do with it...
Re: (Score:3, Funny)
Re: (Score:1)
Or perhaps joke fail. I sure didn't think it was hilarious.
Re: (Score:2)
Or perhaps joke fail. I sure didn't think it was hilarious.
Your response is a common and erroneous defensive response.
Just because you did not find a joke to be funny does not mean it was not still a joke. Your post that marvel comics only has one 'L' was not a criticism of the humor in the joke, it was an attempt to correct a perceived error. If you knew it was a joke you would have known the error was intentional and would not have attempted to correct an error that was not really an error at all.
Re: (Score:2)
s/you/he/g
Re: (Score:2)
You missed his joke!
He didn't say it wasn't a joke, just that it was a bad one.
Re: (Score:1)
He didn't say it wasn't a joke, just that it was a bad one.
Context.
Re:Faulty Marvell Chips, eh? Wha... u jusst cain't (Score:1)
get no SATAs...faction? Awwww...
Re: (Score:2, Funny)
6G a stop-gap solution for high-end SSDs, anyway (Score:4, Insightful)
Even the 6G standard won't hold for high-end SSDs (which seem to be raid striped in one unit, AFAIK). The long-term solution for those are ones that connect via PCIe, so this doesn't seem to be that big a deal, really.
Re: (Score:3, Interesting)
The PCIe versions of SSDs I've see so far doesn't seem to support booting. That's a major crimp in their usefulness. Also, they're not all that useful in laptops.
Re: (Score:3, Interesting)
The PCIe versions of SSDs I've see so far doesn't seem to support booting. That's a major crimp in their usefulness.
True, but that's temporary. I've heard the next major crop of them coming out over the next quarter or so will be bootable. I certainly wouldn't buy one until they ARE bootable.
Also, they're not all that useful in laptops.
Laptops have PCIe connections available, too; that shouldn't be a problem.
Re: (Score:3)
I'll bet they implemented striping a different way than RAID. Since all they need to do is act like a standard hard drive to the machine, they can implement something faster and more specific to flash memory (since every optimization counts).
Re: (Score:2)
Nitpick:
I'll bet they implemented striping a different way than RAID. Since all they need to do is act like a standard hard drive to the machine, they can implement something faster and more specific to flash memory (since every optimization counts).
Pedantic-Man(tm) Approved(tm)! :)
Re: (Score:2)
Are we really all the way back to hardcards?
I guess fads really do go in cycles.
This is surprising (Score:2, Insightful)
To be bad enough that these low-end board shops would reject it this Marvell part must be truly heinous. Asus et al. usually don't hesitate to ship boards with badly flawed components. Cox talked about this [kerneltrap.org] a few years ago.
If you want good SATA avoid the third party chips these board makers integrate. Especially the RAID crap. Wait for Intel to build it into their regular chipsets.
Re: (Score:2)
I agree Marvell is garbage but it's not like "high-end" board shops (whoever you think those are) are above including shitty parts. My Intel desktop board (D975XBX2)'s primary SATA controller is Intel, the second group of four is powered by some Marvell piece of shit though... 88SE6145 I think. The 4 drives on the Intel chip have no problems. The 4 drives on the Marvell chip randomly reset and have generally poor performance.
nVidia nForce 4, Silicon Image, and JMicron (Score:1)
I had a very bad experience with my old nVidia nForce4 Chipset [nvidia.com] motherboard RAID chip while using it and just recently found that none of the old or new drivers work correctly when the Intel X25-M 80GB SSD is plugged into the motherboard causing my Windows OS to freeze during boot-up when the driver is initialized or the RAID capability just doesn't work at all. I even wrote up an entire account of this problem in a few threads, one on nVidia's forum and another on HardOCP Forum to warn users about trying t
Marvell == suckage in Si form (Score:4, Insightful)
Honestly, Marvell chips have cost me more grief on Linux installs than all other vendors combined. If this gets mobo vendors to design out Marvell, then I say: "Grand!".
Re: (Score:2)
Honestly, Marvell chips have cost me more grief on Linux installs than all other vendors combined. If this gets mobo vendors to design out Marvell, then I say: "Grand!".
The linux gigabit ethernet driver bites.
The current one - sky2 - frequently dies under heavy loads and seems to rely on an internal watchdog timer to kickstart it back to life within 60 seconds or so, making a gigabit nic no faster than a 10mbps nic on average.
The Marvell provided one - sk98lin - worked until recent kernels changed a couple of structure definitions - I manually tweaked it myself to get it up and running, but jumbo frames don't work (turn them on and it seems to lose any packet larger than t
"Me too", not just Linux. (Score:3, Insightful)
I have to agree there. Marvell chips have given me nothing but grief, both in Windows and Linux, both network and ATA. They made me buy a SATA DVD writer to replace the fully functional PATA device, as well as a Silicon Image-based SATA controller to attach harddisks (couldn't use the Intel controller, because MSI still insists on placing the "good" SATA connectors under the GPU cooler).
All in all, I'm happy about anything bad happening to Marvell.
Powered eSATA? (Score:2)
Anyone know if/when powered eSATA is supposed to come online? As in power and data in the same single cable?
It feels somewhat silly that I have two options with my 2½ eSATA/USB:
USB - single cable, low speed
eSATA - two cables, high speed
Might even result in some interesting new types of flash storage devices. HD speeds in USB key sizes.
Ho hum (Score:1)
Wait, that's never going to happen. By the time 6Gbit SATA takes off, SAS will be at 12Gbit.
Not that I'm dissing SATA -- It's cool to be able to connect a boatload of SATA disks to a SAS expander for a (relatively) cheap SAN with half a petabyte of storage.
Re: (Score:2)
I think it's best that you remain asleep.
"a new connector to support 1.8-inch drives" (Score:1)
For the love of heaven, WHY?
Re: (Score:2)
Probably because the SATA power connector plus SATA data connectors are so physically large, that you can get the PATA drives in a smaller form factor? (PATA drives use a 44-pin ZIF socket that's really small and low profile. It's also a touch fragile...).
I wish I could find SSDs easily in a PATA connector - 1.8" drives usually have PATA interfaces and I'd like to upgrade one of my computers to SSD. BUt the Intel ones are SATA only, and
Re: (Score:2)
All you'd need is a different adapter to use the small-connector drives in existing desktop PCs. Changing the data connector seems totally uncalled for.
Marvell (Score:1)