Subject: General Tech | July 6, 2017 - 12:33 PM | Jeremy Hellstrom
Tagged: ram, micron, rumour, SK Hynix, toshiba
As is tradition, after we received hopeful news yesterday about Samsung's investing in the expansion of their flash production we now have bad news out of Micron. DRAMeXchange reported a nitrogen leak in Micron's Taoyuan fab which prompted an evacuation and the possible stillbirth of ~60,000 wafer starts, or about 5.5% cut in the amount of RAM available by the end of the month. Trendforce also reported the same incident and numbers.
Micron has released a statement contradicting these stories, stating that while there was an incident, there was no real impact to the business or to employees. One hopes that is the more accurate report as that particular Fab produces LPDDR4, which is already in high demand and short supply. Indeed, another story mentions that SK Hynix and Toshiba's 3D NAND production was well below expectations and that the supply of NAND for iPhones may fall short by as much as 30%.
This would imply that any impact on Micron's RAM production, even if nowhere near the amount mentioned by the press, would have a large effect on the market in the coming quarters. Samsung will certainly try to capture some of this demand, but the upgrades to their Fabs are still a while off and they are already operating at close to maximum capacity. Fingers crossed we don't hear bad news from GLOFO tomorrow morning!
"Micron Technology has issued a statement regarding recent reports about its fabrication facility in Taoyuan, Taiwan. Micron clarified that there was no nitrogen leaking incident nor evacuation of personnel. A minor event did occurred at the facility, but operations are recovering speedily without material impact to the business."
Here is some more Tech News from around the web:
- John McAfee and Intel settle name battle @ The Inquirer
- Microsoft's redesign of Skype is really upsetting almost everyone @ The Inquirer
- U wot M8? Oracle chip designers quietly work on new SPARC CPU @ The Register
- OpenBSD Will Get Unique Kernels On Each Reboot @ Slashdot
- Create a user called '0day', get bonus root privs – thanks, Systemd! @ The Register
- Cha-ching! NotPetya hackers cash out – but victims won't ever see that data again @ The Register
Subject: Storage | August 11, 2016 - 12:06 PM | Allyn Malventano
Tagged: FMS, FMS 2016, XPoint, micron, QuantX, nand, ram
Earlier this week, Micron launched their QuantX branding for XPoint devices, as well as giving us some good detail on expected IOPS performance of solutions containing these new parts:
Thanks to the very low latency of XPoint, the QuantX solution sees very high IOPS performance at a very low queue depth, and the random performance very quickly scales to fully saturate PCIe 3.0 x4 with only four queued commands. Micron's own 9100 MAX SSD (reviewed here), requires QD=256 (64x increase) just to come close to this level of performance! At that same presentation, a PCIe 3.0 x8 QuantX device was able to double that throughput at QD=8, but what are these things going to look like?
The real answer is just like modern day SSDs, but for the time being, we have the prototype unit pictured above. This is essentially an FPGA development board that Micron is using to prototype potential controller designs. Dedicated ASICs based on the final designs may be faster, but those take a while to ramp up volume production.
So there it is, in the flesh, nicely packaged and installed on a complete SSD. Sure it's a prototype, but Intel has promised we will see XPoint before the end of the year, and I'm excited to see this NAND-to-DRAM performance-gap-filling tech come to the masses!
Subject: Storage, Shows and Expos | September 16, 2014 - 12:49 PM | Allyn Malventano
Tagged: ram, NVMe, IOPS, idf 2014, idf, ddr4, DDR
The Intel Developer Forum was last week, and there were many things to be seen for sure. Mixed in with all of the wearable and miniature technology news, there was a sprinkling of storage goodness. Kicking off the show, we saw new cold storage announcements from both HGST and Western Digital, but that was about it for HDD news, as the growing trend these days is with solid state storage technologies. I'll start with RAM:
First up was ADATA, who were showing off 64GB DDR3 (!) DIMMs:
Next up were various manufacturers pushing DDR4 technology quite far. First was SK Hynix's TSV 128GB DIMMs (covered in much greater depth last week):
Next up is Kingston, who were showing a server chassis equipped with 256GB of DDR4:
If you look closer at the stats, you'll note there is more RAM in this system than flash:
Next up is IDT, who were showing off their LRDIMM technology:
This technology adds special data buffers to the DIMM modules, enabling significantly higher amounts of installed RAM into a single system, with a 1-2 step de-rating of clock speeds as you take capacities to the far extremes. The above server has 768GB of DDR4 installed and running!:
Moving onto flash memory type stuff, Scott covered Intel's new 40 Gbit Ethernet technology last week. At IDF, Intel had a demo showing off some of the potential of these new faster links:
This demo used a custom network stack that allowed a P3700 in a local system to be matched in IOPS by an identical P3700 *being accessed over the network*. Both local and networked storage turned in the same 450k IOPS, with the remote link adding only 8ms of latency. Here's a close-up of one of the SFF-8639 (2.5" PCIe 3.0 x4) SSDs and the 40 Gbit network card above it (low speed fans were installed in these demo systems to keep some air flowing across the cards):
Stepping up the IOPS a bit further, Microsoft was showing off the capabilities of their 'Inbox AHCI driver', shown here driving a pair of P3700's at a total of 1.5 million IOPS:
...for those who want to get their hands on this 'Inbox driver', guess what? You already have it! "Inbox" is Microsoft's way of saying the driver is 'in the box', meaning it comes with Windows 8. Bear in bind you may get better performance with manufacturer specific drivers, but it's still a decent showing for a default driver.
Now for even more IOPS:
Yes, you are reading that correctly. That screen is showing a system running over 11 million IOPS. Think it's RAM? Wrong. This is flash memory pulling those numbers. Remember the 2.5" P3700 from a few pics back? How about 24 of them:
The above photo shows three 2U systems (bottom), which are all connected to a single 2U flash memory chassis (top). The top chassis supports three submodules, each with eight SFF-8639 SSDs. The system, assembled by Newisys, demonstrates just how much high speed flash you can fit within an 8U space. The main reason for connecting three systems to one flash chassis is because it takes those three systems to process the full IOPS capability of 24 low latency NVMe SSDs (that's 96 total lanes of PCIe 3.0!)!
So there you have it, IDF storage tech in a nutshell. More to come as we follow these emerging technologies to their maturity.
Subject: General Tech | March 10, 2014 - 04:50 PM | Jeremy Hellstrom
Tagged: Samsung, micron, Hynix, infineon, nec, toshiba, ram, dirty pool
If you bought RAM between 1998 and 2002 from Samsung, Micron, Hynix, Infineon, NEC, and Toshiba in the USA, you are entitled to a small payout, assuming you have proof of purchase. The DRAM makers never admitted guilt and chose to settle out of court and you have until August 1st to follow the link in The Inquirer's story to put in a claim. If you wish to opt out and sue them yourself you have until May 5th to do so but you might be better off taking the $10.
"Remember getting hosed on those 128MB DIMM RAM sticks back in Y2K? Well, it's time to exact your revenge: with a $10 payout."
Here is some more Tech News from around the web:
- The TR Podcast 151: Key switch explosion, the new System Guide, and overclocked SSDs
- CeBit 2014: MSI Press Conference @ Madshrimps
- Sapphire R9 290X VaporX 8GB @ Cebit
- Microsoft fixes Skype for Windows 8.1 @ The Inquirer
- Enermax unleash 1,700W PSU (peak 1,800W+) PSU @ Cebit
- OS 7.1 arrives on iPhones and iPads with Carplay, Touch ID improvements @ The Inquirer
- Intel Haswell Refresh CPU Details Leaked @ TechARP
- Galaxy S5 vs Galaxy S4 specs comparison @ The Inquirer
- Kinect + Wiper Motor + LEGO = 3D Scanner @ Hack a Day
Subject: Storage, Shows and Expos | January 8, 2014 - 12:57 AM | Allyn Malventano
Tagged: ram, micron, memory, ddr4, CES 2014, CES
While the Crucial did not have much in the way of new flash memory product launches this year, Micron as a whole has been busily churning out further revisions of DDR4 memory. While our visit last year only revealed a single prototype for us to look at, now we have all of the typical form factors covered:
From top down we have enterprise, enthusiast, OEM, and SO-DIMM form factors, all populated with DDR4 parts. All that needs to happen now is for motherboard and portable manufacturers to get on board with the new technology. As with all chicken-and-egg launches, someone needs to take the first plunge, and here we can see Micron has certainly been on the leading edge of things. That enterprise part above is a full 16GB (not bits!) of DDR4 capacity.
Follow all of our coverage of the show at http://pcper.com/ces!
Subject: General Tech, Memory | January 1, 2014 - 03:37 AM | Scott Michaud
Tagged: RDIMM, ram, LRDIMM
In all honesty, outside of on-die graphics solutions, memory speed and latency are often neglected. My only requirements for RAM beyond the recommended specs for my motherboard and processor has been a heat spreader of some sort (and that is just because I have bad luck with several DOAs on unshielded RAM which I assume was handling problems).
But this story is for the enterprise users.
Johan De Gelas of AnandTech decided to test a few different configurations of RAM including both RDIMM and LRDIMM modules. LRDIMMs are significantly more expensive than the cheaper RIMM modules but, especially if you could reduce server count (and active licenses of software running on them) they wanted to investigate whether it could be cheaper overall. This would not be the case if software is completely CPU-limited... but, again, when is memory the limiting factor?
That is where the benchmarks come in. Among the handful of measurements performed, they simulated thousands of users accessing a CDN with between one-to-three-quarters of a terabyte of memory. In both cases, 768GB of LRDIMM memory had significantly higher throughput and significantly lower latency.
As always, check out the review if you are interested.
Subject: Storage | May 10, 2013 - 03:48 PM | Tim Verry
Tagged: ramdisk, ram drive, ram, radeon ramdisk, amd
In light of AMD’s latest memory release and Radeon RAMDisk push, I decided to take a look at the latest version 4.1.0 of the RAMDisk software to see what had changed since the last time I tested it out. Improved installation and logging along with a couple of new features are all part of the new RAMDisk software.
AMD has simplified the installer since the previous version to the point that only a few clicks are necessary to get setup. Although you can jump into the advanced settings and change the installation path, the default options are basically just to accept the ToS and click next. Other GUI tweaks include a new Logging tab that scans the last 1,000 entries in the Windows Event Log and shows only those related to the RAM Drive.
The biggest change is the addition of new options in the load/save tab. Because of the nature of RAM, the RAMDisk created by the software is not persistent across reboots. However, you can save the disk image to a file on persistent storage (a hard drive, SSD, et al). Then, you can save the RAM Drive and its contents to a file and reload that disk after a restart.
The paid version of Radeon RAMDisk takes this a step further by allowing background updating of the RAMDisk data. With the Load in Background option, the RAMDisk will be immediately available to the operating system after a restart. The software will automatically start transferring data from the image stored on the hard drive to the portion of RAM set aside for the RAM disk instead of making the user wait fro the entire disk to be recreated before it can be accessed. Any data requested that has not yet been transferred to the RAM disk will be transparently pulled from the hard drive image.
Further, AMD offers up a background update option that will run in the background and continuously write RAMDisk changes to the *.img file stored on the hard drive. This eliminates the need to wait for the entire RAMDisk to be written to disk before shutting down the computer or stopping the RAM Drive. Considering the wait times to read and write data from/to the hard drive is one of the major limitations of RAM drives, this is a really useful feature that certainly adds some incentive to springing for the paid version.
The free version doesn’t get background updating, but it does still have the AutoSave feature that will write data out to the image file periodically which will help prevent data loss due to power failure or kernel panic.
Heh, the SSD is pegged but the RAMDisk utilization peaked at 4% when copying a 1.51GB Kerbal Space Program (with a few mods installed) folder from an Intel X25-M to a 4GB RAMDisk ;).
In my brief testing yesterday, I had some trouble getting the software to create a FAT32 formatted disk, where it kept changing to unformatted before creating the disk. Eventually I opted to format the drive myself using Windows’ Disk Management utility. Aside from that hiccup, I think the new version is worth updating to if you have not already--especially if you have the paid version (so that you can get the background data transfer features).
For specific details on exactly what has changed, an AMD-provided change log is below:
Feature Highlights of AMD Radeon™ RAMDisk release 4.1
- Updated GUI improvements .NET
- Updated installer package – Fewer clicks required to install
- Improved GUI event logging
- Improved management of options when setting Load/Save
Performance Highlights of AMD Radeon™ RAMDisk release 4.1
- Performance gains on AMD Radeon™ RAMDisk 32GB and 64GB
- Vastly improved load and save mechanics allowing for background update and background loading of the RAMDisk. Reduces wait times for load and save. “Background Update” and “Load in Background” enabled (registered users only)
- Faster PC startup and shutdown while RAMDisk is enabled.
Improved IO performance on multi-processors and multi-core systems
- Evenly distributed load among the CPUs. Allows for more system efficiency.
Subject: General Tech | February 28, 2012 - 01:12 PM | Jeremy Hellstrom
Tagged: ram, optical ram, o-RAM
If this is truly a breakthrough and not just another step towards optical RAM, then this story at The Register should put smiles on a lot of faces. Utilizing an indium gallium arsenide strip buried in gallium arsenide these researchers have found a way to make this material either transparent or opaque to light, and using that to dynamically record a 0 or 1. This would imply that the speed of the RAM would only be limited by how quickly the material can switch from one state to the other. These theoretical DIMMs would need a minute amount of power, the abstract states 30 nW as the predicted power draw, which could eliminate heatspreaders as we know them today. The security conscious may also be assuaged that the state will last for a matter of microseconds so there will be no more freezing RAM for snooping through later.
"Japanese researchers are claiming a breakthrough in all-optical memory, one of the key bottlenecks remaining in the optical communications world.
The high throughput of optical communications systems brings its own problem: any function that can’t be performed in the optical domain demands an opto-electric conversion, creating a bottleneck in the system. This has put a premium on research into optical switching, amplification and signal regeneration."
Here is some more Tech News from around the web:
- Samsung shows off 20nm PRAM @ SemiAccurate
- DDR4 shows up in the wild @ SemiAccurate
- Mutant heroes to rescue Ultrabooks from price trap @ The Register
- Dell Joins Rival HP on Growing List of ARM Server Makers @ DailyTech
- Intel Sandy Bridge RC6 Is Good To Go @ Phoronix
- This image contains a hidden audio track @ Hack a Day
- MWC 2012: NVIDIA Samples Tegra 3+, Tegra 4 to OEM Partners @ VR-Zone
- AMD Heaven GamExperience London Event Gallery @ HardwareHeaven
Subject: Memory | January 9, 2012 - 09:00 PM | Tim Verry
Tagged: ram, memory, ddr3, crucial, ballistix
Crucial, a company most well known for their RAM modules, today announced three new series of Ballistix DDR3 RAM modules. The RAM is available in low latency modules based on Micron's 4 Gb chips, and runs at either 1600 MHz and 1866 MHz. The three new series are called Ballistix Sport, Ballistix Tactical, and Ballistix Elite.
The Ballistix Sport modules are the low end modules of the three new series and are designed for mainstream users and a gamers on a budget. They are available in single, dual, and triple channel matched kits. The single modules are available in a DDR2 module running at 800 MHz, DDR3 stick running at 1333 MHz, or DDR3 DIMMs running at 1600 MHz. The DDR2 DIMM need 1.8 volts and delivers a CAS latency of 5-5-5-15 while the DDR3 DIMMs need 1.5 volts and have a CAS latency of 9-9-9-24. The dual and triple channel kits have the same specifications as the single module DDR3 RAM, though obviously they come with multiple matches DIMMs in one package.
Ballistix Tactical brings up the middle ground in the new lineup, and are comprised of DDR3 DIMMs only. The single DIMMs are available in 2 GB, 4 GB, and 8 GB capacities. They need 1.5 volts, run at 1333 MHz and 1600 MHz, and have a CAS latency of 7-7-7-24 or 8-8-8-24. Like the other kits, they come in single, dual, and triple channel kits. The Dual channel kits come in 4 GB, 8 GB, and 16 GB capacities and the triple channel kits come in 6 GB and 12 GB capacities. Other than the additional sticks of RAM, they run at the same voltages and CAS latencies.
Last up is the top tier of the three new Ballistix series, dubbed the Ballistix Elite. These modules are designed for high performance gaming and memory intensive tasks. They have the most flair as well, with tall aluminum heat spreaders. The Ballistix DIMMs come in single, double, and triple channel memory. The single modules come in 2 GB, 4 GB, and 8 GB capacities. They operate at 1.5 or 1.65 volts and either 1600 MHz, 1866 MHz, or 2133 MHz. CAS latencies vary between the various SKUs and include CAS 8-8-8-24, 9-9-9-27, and 9-10-9-27 (for the module running at 2133 MHz).
The Crucial 8 GB Ballistix DIMMs are able to be installed in configurations up to 64 GB in the case of the Intel X79 motherboards. They are available for purchase now worldwide and are backed by a lifetime warranty. To give you an idea of pricing, the 4 GB Ballistix Sport kit running at 1600 MHz is $33.99 USD while the 8 GB Ballistix Tactical kit running at 1866 MHz is $79.99 USD. Finally, the 8 GB Ballistix Elite kit at 1866 MHz is $87.99 USD.
Subject: Memory | December 23, 2011 - 04:10 PM | Tim Verry
Tagged: supply, ram, price increase, nand, dram market, adata
Computer enthusiasts and OEMs alike have been living the dream of extremely cheap RAM modules; however, Adata CEO Simon Chen believes that the dream may be close to ending. In 2012, the DRAM manufacturers will start to cut production such that they are reducing supply and thus can charge more than they currently can (they have been producing DRAM consistently over the past couple years such that there has been more than enough supply and thus a lower cost). After the holiday season, PC OEMs will start to replenish their inventories and when they do, they will be increasing inventories to a months supply instead of a two week supply.
Chen notes that the four major manufacturers of DRAM chips including Elpida Memory, Hynix Semiconductor, Micron Technology, and Powerchip Technology have suffered from selling the chips at such reduced prices for so long. While DRAM chips produced on older manufacturing processes may still be sold below the cost of production, newer DRAM manufactured on the 30nm process "will rebound from the current bottom level to a level above cash-flow production cost."
In addition to the reduced production and newer process, the demand for DRAM in general is expected to decrease due to the rising popularity of mobile computers, Chen notes. Further, the decrease in desktop DRAM demand is balanced out by increased demand for server memory from data centers purchasing additional RAM direct from the manufacturers as the server OEMs charge a hefty premium for RAM. Due to the shake up in the industry, "many makers of DRAM modules have shifted business operation to other areas" like ruggedized memory and to producing NAND flash chips for SSDs.
Admittedly, the memory makers are walking a fine line between spinning down production and being accused of price fixing; however, the ride has been a good one for consumers for a while now and the manufacturers are likely getting tired of the razor thing profit margins. Chen's analysis of the situation may be correct in light of that fact, the new process technology allowing for better yields combined with generally lower production while the big OEMs will be buying up more RAM for their own inventories may well spell the end of being able to impulse buy tons of DDR3 RAM! What are your thoughts on both Chen's analysis of the price increase and the industry itself- do you think prices are likely to go up next year?