Micron Pushes GDDR5X To 16Gbps, Expects To Launch GDDR6 In Early 2018

Subject: Memory | June 7, 2017 - 01:02 AM |
Tagged: micron, gddr6, gddr5x

JEDEC made the GDDR5X memory standard official almost a year and a half ago where it launched at 10 Gbps and quickly hit 12 Gbps. Set to bridge the gap between GDDR5 and the upcoming GDDR6, the “G5X” standard is quickly catching up to and matching the speeds that GDDR6 will run at.

Specifically, Micron’s Graphics Design Team in Munich was able to achieve an impressive 16 Gbps in their high speed test environment. The team was able to hit 16 Gbps on a “meaningful sampling” of its mass production GDDR5X silicon which makes the feat much more impressive as it means these higher speeds are moving closer to reality than theory. Micron measured a PRBS11 (psuedorandom binary sequence) pattern read at 16 Gbps using an oscilloscope and also showed off a chart that compared the stable data rate timing margin versus data rate from 10 Gbps to 16 Gbps.

Micron GDDR5X.png

In addition to teasing the 16 Gbps memory speed (it will be awhile yet before we see products like graphics cards running memory at those speeds), Micron announced that it expects to being mass productions of GDDR6 chips in early 2018. GDDR6 will see a new (larger) FBGA1180 package, faster base sort speeds (GDDR6 will start at 12Gbps vs G5X's 10Gbps), and moving to a dual channel approach with channels that will have half as many I/O links (GDDR5X is x16/x32 while GDDR6 will be x8/16 per channel). It will be interesting to see how this move will stack up to G5X, but in theory Micron will be able to push clocks even higher (maybe even higher than 16 Gbps) by having more but simpler channels (and it may be easier for graphics card manufacturers to wire up their cards to the memory chips.

SK Hynix, who showed off its first GDDR6 chip at GTC, appears to be following the same I/O design as Micron with two channel memory at x8 or x16 per channel.

Are you ready for faster GDDR5X? Hopefully these new faster G5X chips come out soon to give AMD and NVIDIA a more appealing alternative to HBM and HBM2 for mid-range and high end consumer graphics cards since High Bandwidth Memory seems to still be suffering from limited supply and is holding the GPU guys back on being able to crank up the production lines!

Also read:

Source: Micron

Introduction, How PCM Works, Reading, Writing, and Tweaks

I’ve seen a bit of flawed logic floating around related to discussions about 3D XPoint technology. Some are directly comparing the cost per die to NAND flash (you can’t - 3D XPoint likely has fewer fab steps than NAND - especially when compared with 3D NAND). Others are repeating a bunch of terminology and element names without taking the time to actually explain how it works, and far too many folks out there can't even pronounce it correctly (it's spoken 'cross-point'). My plan is to address as much of the confusion as I can with this article, and I hope you walk away understanding how XPoint and its underlying technologies (most likely) work. While we do not have absolute confirmation of the precise material compositions, there is a significant amount of evidence pointing to one particular set of technologies. With Optane Memory now out in the wild and purchasable by folks wielding electron microscopes and mass spectrometers, I have seen enough additional information come across to assume XPoint is, in fact, PCM based.

XPoint.png

XPoint memory. Note the shape of the cell/selector structure. This will be significant later.

While we were initially told at the XPoint announcement event Q&A that the technology was not phase change based, there is overwhelming evidence to the contrary, and it is likely that Intel did not want to let the cat out of the bag too early. The funny thing about that is that both Intel and Micron were briefing on PCM-based memory developments five years earlier, and nearly everything about those briefings lines up perfectly with what appears to have ended up in the XPoint that we have today.

comparison.png

Some die-level performance characteristics of various memory types. source

The above figures were sourced from a 2011 paper and may be a bit dated, but they do a good job putting some actual numbers with the die-level performance of the various solid state memory technologies. We can also see where the ~1000x speed and ~1000x endurance comparisons with XPoint to NAND Flash came from. Now, of course, those performance characteristics do not directly translate to the performance of a complete SSD package containing those dies. Controller overhead and management must take their respective cuts, as is shown with the performance of the first generation XPoint SSD we saw come out of Intel:

gap.png

The ‘bridging the gap’ Latency Percentile graph from our Intel SSD DC P4800X review.
(The P4800X comes in at 10us above).

There have been a few very vocal folks out there chanting 'not good enough', without the basic understanding that the first publicly available iteration of a new technology never represents its ultimate performance capabilities. It took NAND flash decades to make it into usable SSDs, and another decade before climbing to the performance levels we enjoy today. Time will tell if this holds true for XPoint, but given Micron's demos and our own observed performance of Intel's P4800X and Optane Memory SSDs, I'd argue that it is most certainly off to a good start!

XPoint Die.jpg

A 3D XPoint die, submitted for your viewing pleasure (click for larger version).

You want to know how this stuff works, right? Read on to find out!

Micron Launches SolidScale Platform Architecture, Consolidates NVMe in the Datacenter

Subject: Storage | May 24, 2017 - 08:45 PM |
Tagged: SolidScale, NVMf, NVMe, micron, fabric, Cassandra

A few weeks back, I was briefed on Micron’s new SolidScale Architecture. This is essentially Micron’s off-the-shelf solution that ties together a few different technologies in an attempt to consolidate large pools of NVMe storage into a central location that can then be efficiently segmented and distributed among peers and clients across the network.

Traditionally it has been difficult to effectively utilize large numbers of SSDs in a single server. The combined IOPS capabilities of multiple high-performance PCIe SSDs can quickly saturate the available CPU cores of the server due to kernel/OS IO overhead incurred with each request. As a result, a flash-based network server would be bottlenecked by the server CPU during high IOPS workloads. There is a solution to this, and it’s simpler than you might think: Bypass the CPU!

blog-image-c.jpg

Read on for our deeper dive into Micron's SolidScale technology!

Source: Micron

Crucial expands their MX300 line of SSDs all the way up to 2TB

Subject: Storage | February 14, 2017 - 06:51 PM |
Tagged: tlc, slc, MX300, micron, imft, Dynamic Write Acceleration, DWA, crucial, 3DNAND, 3d nand

Last June Al took a look at the Crucial MX300 750GB and its ability to switch its cache dynamically from TLC to SLC, helping Crucial improve how they implemented this feature along the way.  It proved to be a great value for the money; not the best performing drive but among the least expensive on the market.  Crucial has since expanded the lineup and Hardware Canucks took a look at the 2TB model.  This model has more than just a larger pool of NAND, the RAM cache has been doubled up to 1GB and the dynamic cache has more space to work in as well.  Take a look at this economy sized drive in their full review.

board_lg.jpg

"Crucial's newest MX300 series continues to roll on with a new 2TB version. This SSD may be one of the best when it comes to performance, price and capacity all combined into one package."

Here are some more Storage reviews from around the web:

Storage

Micron Planning To Launch GDDR6 Graphics Memory In 2017

Subject: Graphics Cards | February 4, 2017 - 03:29 PM |
Tagged: micron, graphics memory, gddr6

This year is shaping up to be a good year for memory with the promise of 3D XPoint (Intel/Micron), HBM2 (SK Hynix and Samsung), and now GDDR6 graphics memory from Micron launching this year. While GDDR6 was originally planned to be launched next year, Micron recently announced its intentions to start producing the memory chips by the later half of 2017 which would put it much earlier than previously expected.

Micron Logo.png

Computer World reports that Micron is citing the rise of e-sports and gaming driving the computer market that now sees three year upgrade cycles rather than five year cycles (I am not sure how accurate that is, however as it seems like PCs are actually lasting longer between upgrade as far as relevance but i digress) as the primary reason for shifting GDDR6 production into high gear and moving up the launch window. The company expects the e-sports market to grow to 500 million fans by 2020, and it is a growing market that Micron wants to stay relevant in.

If you missed our previous coverage, GDDR6 is the successor to GDDR5 and offers twice the bandwidth at 16 Gb/s (gigabits per second) per die. It is also faster than GDDR5X (12 Gb/s) and uses 20% less power which the gaming laptop market will appreciate. HBM2 still holds the bandwidth crown though as it offers 256 GB/s per stack and up to 1TB/s with four stacks connected to a GPU on package.

As such, High Bandwidth Memory (HBM2 and then HBM3) will power the high end gaming and professional graphics cards while GDDR6 will become the memory used for mid range cards and GDDR5X (which is actually capable of going faster but will likely not be pushed much past 12 Gbps after all if GDDR6 does come out this soon) will replace GDDR5 on most if not all of the lower end products.

I am not sure if Micron’s reasoning of e-sports, faster upgrade cycles, and VR being the motivating factor(s) to ramping up production early is sound or not, but I will certainly take the faster memory coming out sooner rather than later! Depending on exactly when in 2017 the chips start rolling off the fabs, we could see graphics cards using the new memory technology as soon as early 2018 (just in time for CES announcements? oh boy I can see the PR flooding in already! hehe).

Will Samsung change course as well and try for a 2017 release for its GDDR6 memory as well?

Are you ready for GDDR6?

Podcast #435 - Qualcomm aptX, FSP Twin 500w PSU, Micro 5100 Enteprise SSDs, AMD Fiscal Results, ASUS Tinker Board, ZeniMax

Subject: Editorial | February 2, 2017 - 10:34 AM |
Tagged: podcast, zenimax, UHD Blu-Ray, toshiba, tinker board, Reundant PSU, qualcomm, micron, Laser Networking, fsp, enterprise ssd, DirectX, delidding, asus, aptX, amd

PC Perspective Podcast #435 - 02/02/17

Join us this week as we discuss Qualcomm aptX, FSP Reundant PSUs, Micron Enterprise SSDs, 5G LTE, AMD Fiscal Year, ZeniMax lawsuit, and more!

You can subscribe to us through iTunes and you can still access it directly through the RSS page HERE.

The URL for the podcast is: http://pcper.com/podcast - Share with your friends!

Hosts: Allyn Malventano, Ken Addison, Josh Walrath, Jermey Hellstrom, Sebastian Peak

Program length: 1:46:22

Podcast topics of discussion:
  1. Week in Review:
  2. News items of interest:
  3. Hardware/Software Picks of the Week
  4. Closing/outro
 

Source:
Subject: Storage
Manufacturer: Micron

Introduction, Specifications and Packaging

Introduction:

Micron paper launched their 5100 Series Enterprise SATA SSD lineup early last month. The new line promised many sought after features for such a part, namely high performance, high-performance consistency, high capacities, and relatively low cost/GB (thanks to IMFT 3D NAND which is now well into volume production since launching nearly two years ago). The highs and lows I just rattled off are not only good for enterprise, they are good for general consumers as well. Enterprises deal in large SSD orders, which translates to increased production and ultimately a reduction in the production cost of the raw NAND that also goes into client SSDs and other storage devices.

170130-192150.jpg

The 5100 Series comes in three tiers and multiple capacities per tier (with even more launching over the next few months). Micron sampled us a 2TB 'ECO' model and a 1TB 'MAX'. The former is optimized more for read intensive workloads, while the latter is designed to take a continuous random write beating.

I'll be trying out some new QoS tests in this review, with plans to expand out with comparisons in future pieces. This review will stand as a detailed performance verification of these two parts - something we are uniquely equipped to accomplish.

Read on for our full review of the Micron 5100 MAX 960GB and 5100 ECO 1920GB Enterprise SATA SSDs!

Micron Launches 5100 Series Enterprise SSDs - 3D TLC up to 8TB!

Subject: Storage | December 5, 2016 - 02:48 PM |
Tagged: tlc, ssd, sata, micron

Today Micron initiated the first of a multi-tier launch of a new SATA Enterprise SSD lineup built around their IMFT 32-layer 3D NAND Flash. It may seem odd for a full enterprise line to use IMFT 3D TLC, as that flash has not been known for the high random IOPS demands of the datacenter, but Micron looks to be making it work, and work well.

consistency.png

Above is a performance consistency plot of their MAX model. While this does have the highest OP of all of the models, the consistency is surpassing even NVMe models (using a bus *much* faster than SATA). Sure the results are only using 1-second averages and not our Latency Percentile, but we will be able to pick out any single-IO inconsistencies once we get samples in for detailed review.

iops.png

Saturated IOPS performance also looks good 'on paper'.

The advantage to operating their flash in TLC mode is that the per die capacity moves from 32GB to 48GB, ultimately driving down the cost/GB of these products and making them an easier sell to enterprise customers. It also enables high capacities - the max capacity of the model with the least overprovisioning (ECO) will reach 8TB in a 2.5" SATA form factor when the last leg of this launch is completed later next year.

The three lines are all using the same controller and base firmware, but with differences in how the dies are laid out with respect to expected performance and endurance.

Below are all of the products being launched. All products use a Marvell 88SS1074 controller at SATA 6Gbit:

  • 5100 ECO
    • 2.5" 7mm: 480, 960, 1920, 3840, 7680 GB
    • M.2 2280: 480, 960, 1920 GB
    • Sequential read/write: 540 / 380-520 MB/s
    • Random read/write: 93k / 9k-31k IOPS
    • Endurance: <=1 DWPD
    • Cost / GB: $0.45 - $0.55
  • 5100 PRO
    • 2.5" 7mm: 240, 480, 960, 1920, 3840 GB
    • M.2 2280: 240, 480, 960, 1920 GB
    • Sequential read/write: 540 / 380-520 MB/s
    • Random read/write: 78 (240GB)-93k / 26k-43k IOPS
    • Endurance: 1-3 DWPD
    • Cost / GB: $0.55 - $0.65
  • 5100 MAX
    • 2.5" 7mm: 240, 480, 960, 1920 GB
    • M.2 2280: (none)
    • Sequential read/write: 540 / 310-520 MB/s
    • Random read/write: 93k / 48k-74k IOPS
    • Endurance: 5 DWPD
    • Cost / GB: $0.65 - $0.75

All models come with Micron 'Flex Capacity', which enables custom *increases* in OverProvisioning. Flex Security enables FIPS 140-2 validated 256-bit AES encryption.

The specs are very good when you consider their performance consistency claims, meaning a 74k IOPS random write rating applies to random writes across the *entire span* of the SSD *at steady state*. Consumer SSD firmware typically chokes with this type of workload, even ones equipped with MLC flash.

We will have more on the 5100 Series from Micron as these products are rolled out and sampled to us for performance review.

Press blast after the break.

Source: Micron

Podcast #423 - Logitech G231, DX12 Multi-GPU, GTX 1070/1080 issues and more!

Subject: General Tech | November 3, 2016 - 10:35 AM |
Tagged: vrm, video, skyrim, qualcomm, prodigy, powercolor, podcast, nxp, multi-gpu, msi, micron, logitech, GTX 1080, gtx 1070, g231, evga, dx12, devil box, deus ex: mankind divided, amd, Alienware 13

PC Perspective Podcast #423 - 11/03/16

Join us this week as we discuss the Logitech Prodigy G231, multi-GPU scaling with DX12, Qualcomm buying NXP, issues with GTX 1070 and 1080 cards and more!

You can subscribe to us through iTunes and you can still access it directly through the RSS page HERE.

The URL for the podcast is: http://pcper.com/podcast - Share with your friends!

Hosts:  Ryan Shrout, Allyn Malventano, Josh Walrath, Jeremy Hellstrom

Program length: 1:10:25

  1. Week in Review:
  2. Today’s episode is brought to you by Harry’s! Use code PCPER at checkout!
  3. News items of interest:
  4. Hardware/Software Picks of the Week
  5. Closing/outro

Subscribe to the PC Perspective YouTube Channel for more videos, reviews and podcasts!!

ADATA' Ultimate SU800 SSD, a new controller and NAND

Subject: Storage | October 28, 2016 - 01:31 PM |
Tagged: adata, Ultimate SU800, 3d nand, micron, silicon motion, SM2258G

ADATA's new entry level SSD is the second to the market which utilizes Micron's 3D NAND and also incorporates the new SM2258G controller from Silicon Motion.  ATTO shows the performance you would expect from a drive in this class, 560MB/s read 512MB/s write for sequential data at 128KB and higher, assuming you do not completely fill the SLC cache.  The SSD Review did not see write performance drop off until they had written 60GB in one shot, the drop is quite dramatic but for most users 60GB writes happen infrequently.  Check out the full review if you are in the market for a value priced SSD.

ADATA-SU800-512GB-Exterior.jpg

"The Ultimate SU800, on the other hand, utilizes a newer Silicon Motion controller and is the second SSD in the market utilizing Micron's 3D TLC NAND. This combination of components has us charting into new waters when it comes to evaluating the performance."

Here are some more Storage reviews from around the web:

Storage