Intel Demos 3D XPoint at IDF Shenzhen

First unveiled mid last year, 3D XPoint has been hailed as the next step for memory. Created in cooperation between Intel and Micron, the new memory technology is expected to serve as both a DRAM and NAND replacement in the future. While we all knew it would be crazy fast compared to NAND, we got the first glimpse of its true speed this week during Intel’s demo at IDF Shenzhen.

Unlike NAND flash which has to be written in pages and erases in blocks, 3D XPoint is addressable on the byte level. This gives much lower overhead and allows for higher speeds, especially in random performance. During the demo, the Optane device was able to reach 1.9GB/s in sequential performance. Furthermore, even when conducting random operations, the drive was able to hit 464,300 4K random IOPS. That’s about 1.9 GB/s as well, something current NAND-based SSDs, even NVMe ones, can’t reach. 3D XPoint seems almost symmetrical when it comes to read/writes and sequential/random performance.

Even with all the demos out, both Intel and Micron have been mum about a specific timeline for 3D XPoint to finally reach market. So far, all we know is that there will be 3D XPoint based SSDs set to arrive sometime this year. Hopefully, that will be sooner rather than later.

ASRock E3V5 WS (Intel C232) Super Alloy Motherboard Review

Introduction, Specifications, and Packaging


Today’s motherboard review lands right in the middle of everything. Not because it’s all over the place, but because it has a lot of usage scenarios and support for pretty much any setup. You could use ASRock’s E5V5 WS motherboard in a gaming system as well as in a server setup as both of those setups are fully supported, but its real area of operation is to act as the base for a powerful workstation.

The ASRock E5V5 WS uses an LGA 1151 socket and the board is compatible with both E3-1200 v5 processors such as the one I’ll be using later on in this review, but it also supports normal 6th Generation Intel Core processors – so basically the entire Skylake and Skylake-S line-up. As backbone behind the CPU that you end up using is the Intel C232 chipset that is perfectly tuned for workstation environments and a motherboard like this. Memory wise you can use normal DDR4 DIMMs as well as ECC UDIMM memory modules with a speed of up to 2133 MHz in the four DIMM slots with 15μ gold contacts for a total of up to 64GB RAM.

The gold contacts in the memory slots aren’t the only quality feature in this motherboard. The E3V5 WS motherboard is part of ASRock’s Super Alloy series and comes with premium 50A power chokes and is made of a high-density glass fabric PCB. It also features ASRock’s full spike protection that protects from sudden surges, lightning, and electrostatic discharges (ESD).

The network connection isn’t controlled by the average low-performance controller either and we find an Intel PHY i219LM controller on this motherboard. The Gigabit Ethernet controller supports Wake-On-LAN, supports 802.3az energy efficient ethernet and PXE. The Ethernet controller is one of the parts that is protected by the Full Spike Protection. The second part with protection is the USB. The ASRock E35 WS features two USB 2.0 ports and four USB 3.0 ports on the rear panel and you can connect another four USB 2.0 and two USB 3.0 ports via headers. All of these are protected by the ASRock Full Spike Protection. The audio is also protected by this and we find a 7.1 channel HD Audio controller, the Realtek ALC892. It features Blu-ray audio support and uses ELNA solid audio capacitors. Last but not least, you also find legacy PS2 connectors for both your mouse and keyboard.

 

Internally we find good expansion options too. Since we are dealing with an Intel-based motherboard, we also get IRST for RAID setups across the six SATA3 6 Gb/s connectors. Supported modes include the normal RAID 0, 1, 5, and 10 modes.

There are two PCIe Gen3 x16 slots that support both Nvidia Quadro and AMD FirePro workstation graphics cards as well as consumer graphics cards. There is also full support for the use of AMD’s CrossfireX technology for multiple graphics card usage. As I mentioned earlier, you can make a great workstation with this motherboard. There are three more PCIe Gen3 x1 slots available for further expansion and neither of them will be blocked by the use of two-slot graphics cards. Even with a CrossfireX setup, you’ll still have access to the three x1 slots.

Quality hardware and components are just one side of the story, we also need a good BIOS and proper software to have the best experinece. The E3V5 WS comes with both of those features too. The UEFI BIOS has the easy EZ mode dashboard besides the advanced setup pages, which contains multiple readings of the system’s current status. You can check the most crucial information of your system, such as CPU speed, DRAM frequency, SATA information, fan speed, etc.

Even better, you can reach the tech support purely from the systems BIOS, as long as the system has an internet connection. You can also install LAN drivers onto your windows installation directly from the BIOS.

The ASRock OMG (Online Management Guard) technology allows you to establish an internet curfew or restrict internet access at specified times for your kids. Stay in control of their surfing times without yelling. This probably isn’t the most used function in a workstation system, but it could be used to shut down any connection attempts outside of office hours. Another useful feature is the USB Key function that can replace the password question when you log into windows. Do it by plugging a USB drive in instead.

The well-known ASRock XFast RAM and XFast LAN are also built into this motherboard. The XFast RAM feature allows you to fully utilize the memory space that otherwise can’t be accessed in 32-bit systems, among others. XFast LAN is a quality of service protocol that prioritizes the important traffic so you don’t have to wait for loading times. ASRock Live Update and APP Shop also help to make the maintenance easier.

Specifications

The specifications are taken directly from the manufacturers product page and can as such be subject to changes in future revisions of the product.

Key Features:

  • ASRock Super Alloy
  • Server-Grade LAN chip support
  • Supports the Intel Xeon E3-1200 v5 Processor & 6th Generation Intel Core Processors (Socket 1151)
  • Supports DDR4 2133 & ECC UDIMM memory modules
  • 2 PCIe 3.0 x16 and 3 PCIe 3.0 x1
  • Supports AMD Quad CrossFireX
  • 7.1 CH HD Audio (Realtek ALC892 Audio Codec), ELNA Audio Caps
  • 6 SATA3 and 6 USB 3.0 (2 Front, 4 Rear)
  • Supports Full Spike Protection, ASRock Live Update & APP Shop
  • Microsoft Windows 10 64-bit, 8.1 64-bit, 7 32-bit, 7 64-bit, Server 2012 R2 64-bit, Server 2012 64-bit, and Server 2008 R2 64-bit support

Packaging and Accessories

A quality motherboard also deserves a great look package, and the E3V5 WS got that. The front makes no secret about the boards power.

On the rear of the box, you will find a detailed photo of the motherboard itself as well as feature highlights and specifications.

Inside the box is a user manual, a driver disk, two SATA cables, and an IO shield besides the motherboard itself.

Micron Prepping MX300 With 3D TLC NAND For April

In addition to launching a new series of enterprise SSDs, Micron has also updated us on their plans for their consumer and OEM client business. While it’s interesting and all to read about enterprise drives, most of you probably won’t need the performance or be able to shell out the big bucks for them. To that end, Micron is planning on a number of releases for their Crucial consumer brand, the big one being the MX300.

With a pedigree that stretches back years from the M4, M500, M550, MX100 and MX200, the MX300 will be the first to utilize TLC. Prior to that, all drives had used the faster and more durable MLC. This won’t be any plain old TLC though as the MX300 will be using Micron’s first 3D TLC NAND. This TLC will come in 384Gbit dies and feature better performance and endurance compared to the planar TLC found in the BX200. No word has been revealed about the controller, but I would expect the Marvell 88SS1074 since that controller is optimized for TLC and the M series have always used Marvell controllers. I would expect the MX300 feature some kind of SLC NAND cache as well.

In addition to the MX300, Micron will also be releasing 2 OEM client SSDs as well. While not targeted to consumers, they still might be available in retail channels. These are the 1100 and 2100 series, both with unknown NAND. We can make some guesses about controllers as one is Marvell M.2 SATA and likely the 88SS1074 and the other is Silicon Motion Industries M.2 PCIe, likely the 2260.

Micron also expected 3D NAND to hit the data center lineups later in the year. There is no timeline yet on when we might see Intel and Micron’s vaunted 3D XPoint. Hopefully, we will be able to bring you some performance numbers for the MX300 once it launches later this month.

Micron Announces Enterprise 9100 and 7100 NVMe PCIe SSDs

NVMe has been hailed as the next big storage protocol supplanting the aging ACHI that many SSDs still use. While NVMe is still slowly making its way into the consumer market, the enterprise segment has been readily moving forwards. In light of this Micron is updating their P420m and P320h lineup of SSDs with the new 9100 and 7100 series of drives with NVMe support.

With the larger model number, the 9100 is the higher performing and more costly model. It comes with a PMC-Sierra controller and boasts a high 27W TDP. The drive comes in both PCIe 3.0 x4 or U.2 formats with peak reads and writes at 3.0 GB/s and 2.0 GB/s respectively. Capacities range from 800GB to 3.2TB. Random read/writes are even more impressive at 750k/160k IOPS.

The Micron 7100 is a more mainstream product, utilizing a  Marvell 88SS1093 controller. This is Marvell’s first NVMe PCIe 3.0 controller and uses Micron’s custom firmware. The form factors and capacities are different too, with M.2 and U.2  peaking at 960GB and 1.92TB respectively. Speeds are respectable as well, peaking at 2.5GB/s and 900MB/s for read and write respectively. Random read/writes are pretty good at 235K/40K for the fastest variant though the large capacities suffer due having to use higher capacity NAND dies, reducing parallelism.

Lastly, Micron has chosen to use their latest 16nm MLC. I suppose their 3D NAND isn’t quite ready for prime time and TLC wouldn’t do for enterprise drives. Micron expected their 3D NAND enterprise SSDs to drop in the second half of 2016. You can find more information at Micron’s press release here.

Intel Announces New 3D NAND SSD Lineup

Intel has had a set of super-fast high-speed SSDs based on the cutting edge 3D NAND flash storage chips in the works for some time now, with the first of these drives having been announced on Thursday. This lineup of SSDs combine both amazing speed and capacities of up to 2TB, which may seem impressive, but some customers have been left disappointed, expecting the drives to be capable of up to 10TB.

The drives announced by Intel that make use of 3D NAND include the SD DC P3320, SSD DC 3520, SSD DC D3700 and D3600, which are all targeted at data centers, workstations, storage arrays and other heavyweight business uses. The drives were designed with the needs of enterprises in mind, developing them with a focus on speed, durability and reliability over sheer capacity, which leaves them a long way behind Samsung’s 15.36TB PM1633a SSD.

Intel’s drives make use of Micron’s 3D NAND chips, which only started being shipped last month. These 3D NAND flash storage chips offer improvements in both speed and density compared to typical flash storage as the storage cells are layered on top of one another, instead of side-by-side. That closeness also allows for far higher transfer speeds. In order to capitalize on these speeds, these SSDs connect to PCI Express 3.0 slots, which allows for higher data rates than typical SATA. This has been shown as the drives are capable of data rates of 365,000 input-output per second for read operations and 22,000 IOPS for writes. Sequentially, the read speed is 1600 megabytes per second, and the write 1400 megabytes per second, making it significantly faster than Intel’s previous SSD, the DC S3510. The DC D3700 and D3600 are even able to make use of multiple PCI-E slots in order to boost the speeds even further, offering as much as a 25% speed increase for some operations.

Unfortunately, both the price and the release date for the drives are yet to be revealed, with the only clue being that they would hit the market sometime in the second quarter of 2016.

Micron Starts Sampling GDDR5X to Customers

Even when much of the excitement about VRAM is coming from HBM2, that technology isn’t quite ready for prime time yet. For now, HBM2 is still a ways away and still a premium product. To hold the line, memory vendors have come up with GDRR5X, a significantly improved version of GDDR5. In what is unquestionably good new, Micron has just started sampling their GDDR5X modules to customers, way ahead of their original summer target.

GDDR5X has been moving along quickly since JEDEC finalized the specifications back in January. It was also only last month that Micron got their first samples back from their fabs to test and validate. This means that GDDR5X was easier to implement than expected and the quality of the initial batch was good enough that there wasn’t much to change in the production process.

Micron will be offering GDRR5X in 1GB and 2GB IC’s, allowing for 8GB and 16GB VRAM GPUs on as narrow as 256bit memory buses. The biggest advantage of GDDR5X is the doubling of bandwidth from 32byte/access to 64byte/access. Combined with higher clock speeds that allow for up to 16Gbps and improved power efficiency, the new memory will be a good match for Pascal and Polaris while we wait for HBM2.

Supermicro X11SAE Workstation (Intel C236) Motherboard Review

Introduction, Specifications, and Packaging


Intel recently released their Greenlow based Skylake series Xeon CPUs and it’s a pleasure to take a look at the first enterprise-grade motherboard built for these processors today. I have Supermicro’s X11SAE motherboard on the test bench which is a standard ATX-sized single CPU board, but one with all the trimmings.

“Supermicro’s new X11 UP workstations, long-life embedded systems and motherboards integrate the latest technologies such as USB 3.1 and M.2 as well as step up performance, density and efficiency to provide a new generation of Green Computing solutions,” said Charles Liang, President and CEO of Supermicro. “Indeed, with Supermicro’s first-to-market integration, advanced engineering and architecture expertise, we deliver the widest range of Skylake-S platforms available to the industry, enabling our customers with exactly the best competitive advantage on the market.”

Supermicro’s X11SAE doesn’t just support the new Greenlow Intel Xeon E3-1200 v5 family CPUs, you can also use 6th Gen Core i7, i5, i3, Pentium, and Celeron series processors. The brain behind the motherboard is the new C236 chipset which comes with a lot of enhancements over the predecessors and Supermicro generally added the newest technologies to this motherboard. Additionally, the motherboard supports up to 64GB DDR4 2133MHz ECC UDIMMs in its 4 sockets. While this motherboard does support ECC and non-ECC modules, it doesn’t support RDIMMS, so make sure you get the right ones.

There are plenty of storage features on this motherboard with the eight native SATA3 6 Gbps ports provided directly by the chipset. The ports support RAID 0, 1, 5, and 10 for that extra spice in your storage setup and the motherboard also has two SuperDOM ports with built-in power. You’ll also find a next-gen PCIe M.2 slot beside the default SATA3 ports, allowing you to get that extra speed. The M.2 slot doesn’t support AHCI modules, but 2242, 2260, and 2280 PCIe modules will run at a great speed thanks to the x4 slot.

Further expansion can be added through the two PCI-E 3.0 x16 and three PCI-E 3.0 x1 slots. The Supermicro X11SAE also features two legacy 5V PCI 32-bit slots for use with older hardware despite the upgrade to the newest platform.

Externally the X11SAE has two Gigabit Ethernet RJ45 LAN ports where one is powered by an Intel i210-AT and the other is powered by an Intel i219LM chip. There’s also a DVI, a DisplayPort, and an HDMI out for use when processors with iGPU are inserted into the motherboard. Further, you’ll find two USB 3.0 ports, two USB 2.0 and two USB 3.1 ports on the rear IO area. The USB 3.1 ports sport the increased bandwidth and power which allows it to run with up to 10Gbps.

Internally you can connect an additionally four USB 3.0 and size USB 2.0 ports. The motherboard also features two COM port headers, an ALC 888S 7.1 HD Audio chip, a TPM 1.2 header, and much more.

A workstation motherboard also needs a good set of monitoring abilities. You can connect up to five PWM fans with status monitor for speed control, on/off settings, and tachometer. The temperature monitoring includes the CPU and chassis environment as well as CPU thermal trip support and I2C temperature sensing logic and Thermal Monitor 2 (TM2) support.

As a Supermicro motherboard, we also have the benefits of their SuperDoctor 5 software that monitors system health of hardware and operating system services from the target nodes in real-time and provides alerts to administrators on the availability of systems in data centers.

Specifications

The specifications are taken directly from the manufacturers product page and can as such be subject to changes in future revisions of the product.

Key Features:

  • Single socket H4 (LGA 1151) supports: Intel Xeon processor E3-1200 v5, Intel 6th Gen. Core i7/i5/i3 series, Intel Celeron and Intel Pentium
  • Intel C236 chipset
  • Up to 64GB Unbuffered ECC/non-ECC, UDIMM DDR4 2133MHz; 4x DIMM slots
  • 2 PCI-E 3.0 x16 (run at 16/NA or 8/8), 3 PCI-E 3.0 x1 (in x4), and 2 5V PCI 32-bit slots
  • Single GbE LAN with Intel i210-AT, Single GbE LAN with Intel i219LM
  • 8x SATA3 (6Gbps) via C236; RAID 0, 1, 5, 10 + 1 PCIe M.2 (support PCIe x4 only, 2242/2260/2280)
  • I/O: 2x COM, TPM 1.2 header
  • 2x SuperDOM with built-in power
  • 6x USB 3.0 (2 rear + 4 via header), 8x USB 2.0 (2 rear + 6 via headers), 2x USB 3.1 (10Gbps, rear)

Packaging and Accessories

The Supermicro X11SAE comes in a neutral package and that’s all that’s needed for a board like this. It isn’t one you’ll find on a shelve in the local store while browsing around, it’s one your order because you know it’s the one you want. There’s still a sticker on the side that will tell you what’s inside and what the base specifications are.

The rear of the box explains a little bit about the series of motherboards which this one is part off. You get basic information about the series as well as a quick view of optimized chassis and systems.

Inside the box we find four SATA3 cables and the IO shield next to the motherboard itself.

Micron Declares GDDR5X Right on Track

Engineers at the Micron Development Center in Munich have announced that they have gotten their first samples of GDDR5X back from their fab before schedule and have started testing. In addition to that, Micron is expecting to ramp up volume production of GDDR5X on their 20nm memory process sometime in mid-2016. GDDR5X is an evolution on GDDR5 rather than a new memory technology and is expected to tide the industry over till HBM2 and HMC come online.

In early testing, some of the GDDR5X samples have already hit 13Gbps, just short of the eventual 14Gbps goal for the production modules. Combined with a new improved prefetch and new quad data rate, GDDR5X is expected to double the bandwidth over GDDR5 while increasing capacity and reducing power consumption. With progress going well, samples will begin to ship to partners (like AMD and Nvidia) in the spring. This means we will be unlikely to see any GDDR5X based cards until fall 2016.

While GDDR5X will still fall short of HBM2 bandwidth, it will undoubtedly be cheaper. It will also allow GPUs to be made with narrower buses while still maintaining the same overall bandwidth, allowing for reduced power consumption, cheaper GPUs and faster GPUs. We can expect the mainstream and even performance segments to utilize GDDR5X while the budget cards stick with GDDR5 and the enthusiast cards use HBM2. For more on GDDR5X, check out our write-up here.

Flash Memory Density Surpasses Traditional Hard Drives

SSDs are most definitely on the rise, being the go-to source of faster boots and the best loading experience for software and games. Traditional disk-based hard drives have always continued to rule the domain of mass storage, having advantages over SSDs in both areal density and price per Gigabyte. One of these advantages is about to become history now, with Micron revealing at the 2016 IEEE International Solid State Circuits Conference that its NAND flash storage had areal density beyond that of hard disk drives.

Micron is far from the only company aiming to improve the performance and density of flash storage. SSD market leader Samsung revealed in 2015 that their flash storage offerings had as much as 1.19Tbits per square inch (Tbpsi), predicting that by 2016, this could reach as much as 1.69Tbsi. Micron’s revelation of their 3D NAND technology had smashed the aims of Samsung, having reportedly demonstrated density as high as 2.77Tbpsi in their laboratory tests. This puts the 3D NAND head-and-shoulders above the densest HDDs around, which only offer around 1.3Tbsi, with many consumer drives offering as little as 850Gbsi despite using techniques like shingled magnetic recording to increase density.

Of course, many of the super-high density flash devices shown at ISSCC are definitely laboratory pieces and not consumer devices, SSDs that are on the market are definitely not far behind. Samsung announced last year that they had plans to release a 16TB SSD and Fixstars already offer a 13TB SSD costing a whopping $13,000. And while price per GB, HDDs still hold an advantage, the tables could be turned as soon as 2017, with the tumbling price of SDDs vastly outstripping the relatively stable price of hard drives.

What this all means is that the future could be bleak for the hard drive. With its main advantages of density and price set to vanish in the next few years, it is likely that we will see more and more SSDs appearing in both desktops and laptops. I for one look forward to the mass adoption of flash memory technology over the traditional hard drive as I doubt many will miss the times waiting for their hard drive to spin up or the painful sounds of their parts failing.

Image credit to Micron

D-Link DXE-820T Dual Port 10Gbase NIC Adapter Review

Introduction


Among our recent upgrade to 10 Gbit networking here in the eTeknix review section wasn’t just the awesome 12-port smart switch that we saw a little while ago, D-Link also supplied me with a DXE-820T dual-port 10 Gbit Ethernet Network card to be able to connect with the switch in full speed. After all, what good does a 10 Gbps switch do when my test rig doesn’t have the same kind of performance.

The D-Link DXE-820T is a dual-port 10 GBASE-T RJ-45 PCI Express high-performance adapter designed for the high-speed PCI Express bus architecture. This adapter offers the increased bandwidth needed in modern environments as well as being a reliable and functional PCI network card. It has been specifically designed to allow throughput at rates up to 40 Gbps, thus eliminating the bottleneck that exists with current 32 and 64-bit PCI bus architectures.

The DXE-820T requires a PCI-E v2 x8 or x16 slot for enough bandwidth, but you don’t need to provide any extra power connectors or other things besides your network cables, naturally. The card is capable of a transmitting distance of up to 100m with Cat 6A or higher in 10 GBASE-T mode and up to 100m with Cat 5 type cables in Gigabit mode. This makes the placement of the switch a lot easier.

The DXE-820T features TCP, UDP, and IP checksum offloading functionality, which transfers the checksum processing tasks from the computer’s own CPU and onto the network card itself. The DXE-820T’s ability to handle the checksum processing means that the CPU’s processing power can be used for other tasks while still achieving 20 Gbps network speeds. It also means that the network card needs some extra cooling in the form of an active fan as well as a passive heatsink.

The adapter also features an onboard screening of 802.1Q VLAN tagged Ethernet frames, allowing you to assign multiple subnets to each server and isolate devices within each VLAN from the rest of the network for better traffic control and security. With support for advanced features such as 802.3x flow control, jumbo frames, and SNMP for network management, the DXE-820T can easily interoperate with your current networking equipment.

With two ports at your disposal, you can increase the network throughput even further than the 10 Gbit per second on each port. With Smart Load BalancingTM, the DXE-820T can configure multiple adapters to work as a team, sharing traffic and ensuring data reliability. This both creates a faster network and provides fault tolerance resulting in a stable and efficient network.

The low profile design allows the deployment in space restricted areas and D-Link also included a low-profile slot cover right away. The card itself is powered by Broadcom’s BCM57810 chip.

The card naturally supports Jumbo Frames for optimized setups and it allows for values up to 9K. The DXE-820T is compatible with all major operating systems with drivers available for both user systems and server systems such as Windows Server 2012 or Solaris 11.

Feature Highlights

  • High-Speed data transmission at rates up to 40Gbps allow for seamless data transfer.
  • 10GBASE-T technology supports distances to 100m over CAT6A or better copper cable.
  • Advanced Features: 802.3x flow control for traffic management, 802.1Q VLAN tagging for increased security, and checksum offloading to reduce CPU processing burden.
  • Bandwidth Management: NIC partitioning enables administrators to manage bandwidth for greater network efficiency.

Specifications

Packaging and Accessories

The DXE-820T NIC card comes in a very neutral box that only really reveals that we have a D-Link product inside. But it’s a network card and it isn’t like you would put the box in the display after installing the card anyway, so simply is good.

We do find a little sticker on the rear of the box that reveals what is inside, the model number along with serial, mac address, and hardware version.

There is both a driver disk and a quick installation manual inside the box, and D-Link also included a low-profile bracket for use in small-form-factor systems. Everything you need to get going with that extra speed.

Micron Forecasts Surprise Loss In Current Quarter

Being one of the biggest DRAM and NAND manufacturers in the market, Micron have chugged along steadily, with the rise of smartphones and tablets helping offset the losses on the PC side. This is set to change very soon though with them forecasting a loss in the current quarter (Q4 2015), the first in a long while. The loss comes due to the naturally weak first quarter of the year, increased investments, pricing pressure and low demand.

Overall, they are expecting to lose between $50 to $120 million, or about 5-12 cents per share. This comes as revenues are expected to fall about $200 million short of analysts expectations. This comes after the company have made a number of acquisitions including Tidal Systems for their SSD controller and Inotera for their DRAM business. Micron is also investing heavily into TLC NAND, 3D XPoint and 3D NAND so it’s not as bad as it first may seem as the investments will hamper their fiscal results, but ultimately should pay off if the investments go as expected.

Coupled with the expected low PC demand, they are also facing heavy pricing pressure from competitors that has led huge price drops for both SSDs and DDR4 over the past few months. This is great news for consumers making DDR4 based systems much more affordable and open to a mini price-war involving Micron and their competitors. With increasing expenses in a time of falling revenues, it’s not surprising that Micron is facing some immediate trouble. The incoming investments though should help bump Micron back into the black soon enough and we will be monitoring this story closely as it develops in the near future.

Do you own any Micron products yourself? The company portfolio includes Crucial for consumer based memory and storage, Micron themselves for enterprise and business class products and Lexar Media who manufacturer flash memory for cameras and recording equipment.

D-Link DXS-1210-12TC 12-Port 10GBASE-T Switch Review

Introduction


As a review site, we need to keep our testing equipment above or at least on the same level as the hardware that we test in order to provide you with relevant reviews and in that regards we’ve just upgraded our network test are with new 10 Gbps equipment. Among the new devices is the D-Link DXS-1210-12TC 10 Gigabit Ethernet Smart Switch that I’m taking a closer look at today.

In order to test this switch I need two systems with 10 Gbps capabilities and four CAT7 cables and below you see all but the one network card as that is onboard in one of my test systems. In my opinion, we have been stuck at Gigabit level for way too long and 10 Gigabit Ethernet is slowly starting to make an entry. It is starting to become more affordable which allows SMBs and enthusiasts to enter an area of the market that previously was limited to enterprise and large corporations.

D-Link’s DXS-1210 series is composed of several 10 Gigabit Ethernet Smart Switches and they are a cost-effective 10 solution capable of servicing a wide range of network needs in any business. Some of the switches in the series focus primarily on SFP+ connections which aren’t optimal for my test scenario. I still test a lot of normal Gigabit equipment and my switch needs to stay compatible to those. Luckily D-Link also has this one with eight 10GBASE-T RJ45 ports and two 10Gbps SFP+ ports. Additionally, since it has two combo ports for use with either RJ45 or SFP+ connections.

The DXS-1210-12TC uses a default 19″ rack format which makes it easy to deploy in server cases and D-Link also included rubber feet in case you should run in placed on a desk or shelve somewhere. The 10GBASE-T copper ports utilize RJ45 interfaces and support 10 Gigabit speeds over CAT6a/CAT7 LAN cabling for up to 100m. That makes it easy to deploy no matter how far apart your systems are. The 10GbE fiber ports can be connected to other 10GbE ports with low-cost Direct Attach Copper (DAC) cables or with SFP+ optical transceivers over multimode or single mode fiber optic cabling.

D-Link made sure that there’s plenty of overhead when it comes to transmission capabilities in the DXS-1210-12TC. It is capable of up to 240Gbps switching capacity, double what is needed, and has a maximum packet forwarding rate of 178.56 Mpps.

A device like this will naturally consume quite a bit of electricity as it has to perform flawlessly at any time and needs the power to do so, but D-Link applied their Green technologies in order to improve on this area without sacrificing operational performance or functionality. The DXS-1210-12TC features smart fans that don’t run at full speed when it isn’t needed, internal heat sensors that monitor and detect temperature changes in order to adjust the fans. The switch can also toggle ports into sleep mode automatically and it is built with hardware parts that in themselves are energy-efficient. The DXS-1210-12TC will still draw between 42.65 watts in standby mode and up to a maximum of 90.81 watts.

Feature wise you’ll find anything you’ll want in a smart switch starting with a complete lineup of L2 features. The DXS-1210-12TC supports port mirroring, Spanning Tree Protocol, and Link Aggregation Control Protocol (LACP). DXS-1210 switches also support Auto Surveillance VLAN (ASV), and Auto Voice VLAN, which are best suited for VoIP and video surveillance deployments. The DSCP markings on Ethernet packets enable different levels of service to be assigned to network traffic. As a result, these voice and video packets take precedence over other packets. In addition, with bandwidth control, network administrators can reserve bandwidth for important functions that require a larger bandwidth or might have high priority.

The DXS-1210 series also features an extensive list of safeguards against traffic flooding and other malicious attacks. It supports 802.1X port and host-based authentication as well as external RADIUS server authentication. It also features Access Control List (ACL) for enhanced network security as well as ARP spoofing prevention. The DHCP server screening feature can also detect rogue DHCP server packets from user ports in order to prevent unauthorized IP assignment.

 

Maintenance is an easy job when it comes to the DXS-1210-12TC smart switch where it also features a few extra functions to make the admins life easier. It includes loopback detection which is used to detect loops created by a specific port and automatically shut down the affected port. There is also a cable diagnostic feature that’s primarily designed to rapidly discover errors and determine a cable’s quality.

The switch has a smart discovery protocol built in which makes it easy to set up and manage through D-Link’s Network Assistant Utility and it will detect all switches within the same L2 network. This allows extensive switch configuration and basic setup of discovered devices, including password changes and firmware upgrades. The switch also supports D-View 7 and Command Line Interface (CLI) through Telnet. D-View 7 is a network management system that allows for the central management of critical network characteristics such as availability, reliability, resilience, and security.

Feature Highlights

  • Green Technology
    • Power saving via the following features: Link Status detection, LED Shut-Off, Port Shut-Off, System Hibernation
  • Security Features
    • Access Control List
    • IP-MAC-Port Binding
    • Clientless MAC/Web access control
    • D-Link Safeguard Engine
    • Port Security
    • ARP Spoofing Prevention
  • Intuitive Management
    • D-Link Network Assistant Utility or Web-based GUI
    • CLI through Telnet
  • Advanced Features
    • Auto Surveillance VLAN
    • Loopback Detection
    • Cable Diagnostics
    • Static Route
    • LLDP/LLDP-MED
    • Auto Voice VLAN

Package and Accessories

Equipment like the D-Link DXS-1210-12TC smart switch doesn’t need a fancy package, it is what is inside that matters.

There is a small sticker in one corner of the box that shows what is inside and the feature highlights.

Inside the box are several leaflets about security and conformity as well as a Getting Started Guide and a Driver and Manual disk.

Included is a mounting system for use as a 19″ rack unit as well as four small rubber feet for those who wish to use it placed on a surface. Screws for the 19″ mounts are included too and so is a cable tie system for the power cord. You wouldn’t want to jank that out by accident.

Naturally there’s also a power cable for the region where you purchased the switch. The switch supports from 110V to 240V, so the only difference here is the plug on one end.

Micron to Acquire Taiwanese Chip Company Inotera For $3.2 Billion

Micron Technology Inc. is an American manufacturer of memory chips, including dynamic random-access memory, flash memory, and solid-state drives. The company announced today it will purchase the remaining 67% of shares in Taiwan’s Inotera Memories Inc. This is a deal valued at $3.2 billion and gives Micron full ownership of this particular outfit. During November, Inotera’s monthly sales fell by 23% and reported the lowest monthly revenue this year. Micron’s Chief Executive, Mark Durcan said in a statement:

“The acquisition is the culmination of a highly successful seven year partnership with Inotera,”

The semiconductor business is very competitive and perhaps Micron wants a greater say in the production line. This is clearly a huge sum of money, but the falling sales figures might have made the takeover bid make more business sense. With the growth of ultra fast PCI-E drives, there’s the possibility of SATA devices being replaced in the future. Although, this could take some time. Whatever the case, it will be fascinating to see how Micron’s development continues and the effect of this acquisition.

Are you using a SATA, M.2 or PCI-E boot drive?

GDDR6 Memory Coming to Graphics Cards in 2016?

The graphics card market is full of interesting power struggles and if recent reports are true, it seems 2016 will be one of the biggest battles yet. AMD may have already put out some cards with HBM memory, and we’ve heard that Nvidia will be doing the same soon too, but don’t count GDDR memory out just yet! It seems that the upcoming GDDR5 standard is being developed by Micron, which will power mid-end graphics cards, while HBM2 will likely remain for higher end cards.

Of course, there’s some confusion here as JEDEC are already working on the GDDR5X standard, so where Micron fits in really remains to be seen, but that’s something we’ll have to wait and see. GDDR5X is said to double the bandwidth, so is GDDR6 is new standard, or just a further refinement of 5X? Either way, we can expect it to adopt a lower node, most likely starting from 20nm and working down from there, allowing for higher clocks, and lower voltages, although these kinds of improvements are the obvious targets for any increase in performance these days.

Our guess is that the revised GDDR standards will be acting as a bridge until HBM matures enough to cover a wider range of cards and budgets. Either way, 2016 is shaping up to be an exciting time in the GPU market, with new memory, new architectures, new cards and so much more on the horizon.

Crucial DDR4 2400MT/s 8Gb-based Server Memory Now Available

Crucial is ready with the next step in their server memory and announced the availability of the Crucial DDR4 2400MT/s 8Gb-based RDIMM, LRDIMM and ECC UDIMM server modules which enable increased performance, bandwidth, and energy efficiency.

The higher density 8Gb-based modules allow for both a greater channel bandwidth and channel density, but the most important factor is probably the lowered power consumption. Memory can be quite power hungry and will make up quite a bit of the overall consumed power in a server environment due to the constant rewrites happening. The new 8Gb-based modules offer up to 20 percent higher energy efficiency than the 4Gb-based modules and that is something that will make a noticeable difference.

Ultimately, these benefits provide more value per gigabit than current 4Gb-based offerings, making it easy to scale
up server deployments in the future and the modules are designed to be compatible with Intel’s next generation processor product families.

Crucial’s 8Gb-based server memory is extensively tested to mission-critical standards and is backed by a limited lifetime warranty. The new 8Gb-based modules are available for immediate purchase through global partners and directly through Crucial.

“We are excited to continue Intel’s collaboration with Crucial with the release of the new 8Gb-based DDR4 server modules,” said Geof Findle, director of memory enabling, Intel. “By working together, we are able to support next-generation server platforms while providing the technology and services needed to support our mutual channel customers.”

“Data-intensive server applications continue to require higher densities of memory as they struggle to meet ever-increasing and more demanding workloads,” said Michael Moreland, worldwide product marketing manager, Crucial. “The new Crucial 8Gb-based server memory modules will help with future scalability and deliver a lower total cost of ownership for users.”

Micron Buys Tidal Systems – Acquires In-House SSD Controller

Right after announcing their 16nm TLC plans, Micron has acquired an SSD controller firm. Tidal Systems, made up of engineers from Link-A-Media Devices (LAMD) and SandForce was founded last year and has managed to create 2 of their own SDD controllers already. With in-house controllers, Micron will be able to iterate more quickly and provide a more unified SSD platform.

For the longest time, Micron has had to rely on third-party controllers to power their SSDs. While Silicon Motion does provide a somewhat complete controller and firmware package, Micron’s main supplier Marvell does not. This has meant that Micron has long had to write their own firmware for most of their SSDs. With the controller now in-house as well, Micron has room to reduce costs and also build on their long expertise in writing firmware.

Tidal Systems also brings their expertise around Low-Density Parity Check (LDPC) to the table. As NAND transitions to use TLC (Triple-Level Cell) and QLC (Quad-Level Cell) on ever smaller lithographies, the number of errors is sure to go up. With LDPC, SSD makers will be able to offset some of the error and endurance issues.

Even with Tidal Systems in-house now though, don’t expect any Micron drives to ship with the new controllers anytime soon. Toshiba took quite a while to integrate their OCZ controllers with their drives. With this purchase, it only leaves SanDisk as a major NAND and SSD producer without their own controller. It is critical though that SanDisk has been able to do the most with Marvell controllers, leveraging their in-house firmware to eek out performance that rival’s Samsung’s.

Thank Tom’s Hardware for providing us with this information 

Micron to start selling TLC SSDs this quarter

Micron Technologies is finally moving into the TLC (Triple-Level Cell) NAND market, with shipments of consumer SDDs starting in Q4 this year. TLC is generally cheaper than MLC, holding up to 3 bits compared to the 2 with the older technology. This allows for cheaper SSDs as it requires fewer NAND dies to reach a certain capacity compared to MLC. By Q3 2016, Micron is expecting about 50% of their SSDs will be using TLC.

Along with Intel, Micron’s joint venture IMFT has largely focused on shrinking the process with MLC in order to gain die savings. IMFT was one of the earliest NAND producers to reach 20 and 16nm. On the other hand, SanDisk/Toshiba and Samsung have long been using TLC in their SSDs, preferring to move to new processes slower. Samsung for instance, released their first TLC  drive back in 2013 and SanDisk their’s in 2014.

Combining their leading 16nm process with TLC should offer great savings for SSD buyers. TLC does come with a number of drawbacks though, most notably lower endurance and performance. TLC generally only can last 1,000 P/E (program/erase) cycles while MLC dos much better at 3,000. Even with only 1,000 cycles though, TLC should be more than enough for most consumers, especially if a bit of extra NAND is set aside. On the performance front, a caching system like those used in the Samsung 840/850 EVO or SanDisk Ultra II can mitigate most of the issues. Micron first trialed their caching system with the MX200 which should be a great starting point to work from for the TLC drives.

Micron has not yet revealed any details for their TLC drives. Intel too, given their joint IMFT venture may be trying out TLC drives in the near future. Combined with 3D XPoint and 3D NAND from SanDisk/Toshiba, 2016 should be a pretty good year for SSDs.

Supermicro X10DAX (Intel C612) Workstation Motherboard Review

Introduction, Specifications, and Packaging


Today I’m taking a look at a motherboard for those that need that extra bit of power behind the curtains for heavy workstation or server tasks. The Supermicro X10DAX is aimed at workstations, but it would work well as a basis for a server as well with the features at hand. In the end, as always, it comes down to what tasks the system has to perform.

Supermicro’s X10DAX is a dual CPU motherboard with plenty of memory and multi-GPU support. The motherboard supports up to 1TB ECC DDR4 2133MHz LRDIMM memory over its 16 DIMM memory banks as well as Triple-SLI setup or Quadro-SLI with dual-GPU graphics cards. Those who will use RDIMMs instead of LRDIMMs will be limited to 512GB maximum memory capacity.

The motherboard supports up to two processors from the Intel Xeon E5-2600 v3 family thanks to the Dual Socket R3 (LGA 2011). That means up to 18 cores and 45MB cache per processor which in return equals a lot of power. The chipset is the Intel C612 which by itself provides a lof of features and enhancements over the predecessor.

One of the unique features on this motherboard versus its brothers is the overclocking features thanks to the BCLK settings in the BIOS and Supermicro has dubbed this Hyper-Speed on their motherboards. This will allow you to get that extra power from your system when it is needed and if it is needed. A great bonus that surely will be appreciated by many of the people purchasing this motherboard. Next to the Hyper-Speed technology, the Supermicro X10DAX also includes Hyper-Turbo mode which allows more power to be delivered to the CPUs and thereby maximize the CPU Turbo Mode frequency.

Supermicro also made sure that the motherboard is Thunderbolt 2 ready, but that’s an optional add-on you’ll have to get. Since this is a workstation board, it also comes with an onboard sound card. It could, for example, get a pretty hard job to edit videos without sound. The Supermicro X10DAX provides onboard 7.1 channel HD audio for just that reason.

Storage connection isn’t a problem either and we don’t have to pay for an extra SAS controller here that we probably don’t need in a workstation. Ten SATA3 ports, of which two supports SuperDOMs, should be plenty for your storage needs. If you need more than that, you’ll most likely have a server located somewhere anyway and thus eliminating the need again. You also get a Type-A USB 2.0 connector on the motherboard, placed right next to the SATA connectors. Since the SATA3 ports are controlled by Intel’s C612 chipset, we have RAID 0, 1, 5, and 10 at our disposal on this board.

Dual Gigabit LAN is included thanks to a couple of Intel i210 network controllers, but we don’t find any dedicated IPMI on this motherboard. The PCI Express connectors available are three PCI-E 3.0 x16, two PCI-E 3.0 x8, and one PCI-E 2.0 x4 (in x8).

With such an amount of power at our disposal, we also need proper cooling. The Supermicro X10DAX has eight 4-pin fan headers that support tachometer monitors, status monitoring, and PWM fans. There are no dedicated CPU fan headers, so you can use any two of the eight for that.

Specifications

The specifications are taken directly from the manufacturers homepage and are as such subject to change in possible future revisions.

Key Features

  • Supports Hyper-Speed
  • Supports 3-way Geforce SLI
  • Supports 4-way SLI support with dual GPU graphics cards
  • Dual socket R3 (LGA 2011) supports Intel Xeon processor E5-2600 v3 family; QPI up to 9.6GT/s
  • Intel C612 chipset
  • 16x DIMM slots for up to 1TB ECC DDR4 2133MHz memory
  • 3 PCI-E 3.0 x16, 2 PCI-E 3.0 x8, and 1 PCI-E 2.0 x4 (in x8) slot
  • Intel i210 Dual port GbE LAN
  • 10x SATA3 (6Gbps) ports (RAID 0, 1, 5, 10)
  • 6x USB 3.0 (4 rear, 2 via header), 5x USB 2.0 (2 rear, 2 via header, 1 Type A)
  • 7.1 HD Audio with optical S/PDIF

Packaging and Accessories

The box for the Supermicro X10DAX is a default layout for the entire line-up and as such presents key features from all boards. While we can spot such things as 40GbE here, we won’t find it on this board.

The rear of the box details the different boards from Supermicro in this category for easy comparison.

Inside the box, next to the motherboard itself, we find six SATA3 cables and an IO shield with proper padding and yet unopened ports. Remember to only pop the ones that the motherboard actually has.

Included is also a Quick Start Guide that will show you the headers and jumper positions, how to install your memory depending on the amount of modules you are using, as well as front header layout and other vital information for installation.

Gigabyte MW70-3S0 (Intel C612) Dual CPU Workstation Motherboard Review

Introduction


Gigabyte’s motherboard series based on the Intel C612 chipset is quite an impressive one and I’ve already had the pleasure to test the MU70-SU0 server motherboard. Today I’m taking a look at its bigger brother, the dual socket Gigabyte MW70-3S0 workstation motherboard.

The Gigabyte MW70-3S0 workstation motherboard is based around Intel’s C612 chipset and offers you two LGA 2011-3 sockets with support for Intel’s Xeon E5-2600 V3 processors as well as 16 DDR4 modules for a total of up to 768GB RAM. The MW70-3S0 is a workstation board and as such it offers some other features than server boards, such as Multi-GPU configuration. With support for up to three graphics cards in CrossfireX or SLI via the three PCIe 3.0 x16 slots and coupled with the CPU and memory support will make sure that we don’t run out of performance anytime soon.

One of the really great things about this board’s memory support is that it supports the full speed of 2133MHz, even when all 12 memory banks are populated. In the past, you had to make the choice, whether you wanted capacity or speed. Gigabyte made sure that you’ll have a maximum speed on all memory banks at all times.

We also get plenty of storage abilities on Gigabyte’s MW70-3S0. Not only does it provide 10 SATA3 ports via the Intel C612 chipset where two of them support SATA DOMs, we also get an LSI SAS 3008 onboard controller with two Mini-SAS ports for eight SAS 12GB/s drives. The SATA ports support RAID 0/1/5/10 while the SAS ports support RAID 0/1/1E/10. The connectivity is secured by two Intel i210 Gigabit Ethernet ports that allow for easy trunking and link aggregation.

Overall, this board is built with quality components that should make sure that everything runs stable and for a long time. It is equipped with IR Digital PWM and IR PowIRstage IC controllers to guarantee a stable operating environment and OS-CON capacitors with a minimum service life of 50.000 hours. The installed chokes are high-end ferrite core chokes that again should help deliver a stable power to both CPU and memory.

Specifications

The specifications are taken directly from the manufacturers homepage and are as such subject to change in possible future revisions.

Key Features

  • Intel Xeon processor E5-1600 V3 & E5-2600 V3 product families
  • 16 x RDIMM/LRDIMM ECC DDR4 DIMM slots
  • Supports 3-Way NVIDIA SLI / AMD CrossFireX
  • 2 x GbE LAN port (Intel I210)
  • 2 x Mini-SAS (for 8 x SAS 12Gb/s ports)
  • 10 x SATA III 6Gb/s
  • 7.1+2 HD Audio
  • 2 x USB 3.0 + 2 x USB 2.0 ports
  • 1 x USB 3.0 header + 1 x USB 2.0 header

Packaging and Accessories

The box follows a very simple design and is perfect for the market it’s aimed at. There is no need for fancy artwork like we’re used to on consumer motherboards.

The rear of the box gives a little detail about 3 key points that are important to a board of this caliber including reliability, availability, and serviceability. These are all major factors for any IT admin needing a reliable product with quality components that is easy to set up and maintain.

Inside the box, we find anything we need to get started, but nothing unnecessary that would drive up the overall price on the product. We get two SATA3 cables where one of them has an angled connector at one end, two CrossFireX bridges, as well as two NVIDIA SLI bridges for 2-way and 3-way SLI and CrossFireX support. There’s also the obligatory IO shield and a driver disk

Intel’s High-Speed SSD Coming in 2016

Intel has unveiled the Optane, its new high-speed SSD brand, at the Intel Developer Conference in San Francisco. The super-fast solid state drive, produced in conjunction with Micron, could be released as early as next year, Intel CEO Brian Krzanich revealed during his keynote speech.

Optane drives operate using Intel’s 3D XPoint technology, which is up to one hundred times faster than NAND. It is non-volatile, has around 10-times greater density than DRAM, but its performance can be anywhere between up to 100 times faster than conventional flash drives, with latencies up to 1000x faster.

However, speeds will be limited for the time being by existing interfaces. According to Intel’s live demonstration during the conference, its early prototype Optane drive has 7.23x the IOPs of Intel’s top-range NAND SSD, the DC P3700, with hopes that the drive will become even faster by the time of release.

“We’re hitting a bottleneck with current storage architecture,” Krzanich explained, adding that computer architectures will need to change in order for users to enjoy the full benefits of 3D XPoint technology.

Optane drives will be available in standard PCIe form, but also in DIMM for Xeon systems, offering greater bandwidth and lower latencies. Both include storage controllers optimised to take full advantage of the 3D XPoint memory.

Thank you Intel for providing us with this information.

ASRock Rack EPC612D4U-2T8R (Intel C612) mATX Server Motherboard Review

Introduction, Specifications, and Packaging


The motherboard that I am taking a look at today is a little bit special as it is considered to be the first mATX motherboard featuring Intel Xeon E5 V3 series with LGA 2011-3 single socket. The ASRock Rack EPC612D4U-2T8R isn’t just a rarely seen form factor in this category, it also packs a huge punch and plenty of features such as quad-channel DDR4 memory support, dual 10 Gigabit Ethernet, dedicated IPMI, and native support for up to 16 storage drives.

Just because you decide to create a motherboard in the mATX form factor as ASRock Rack did here, doesn’t mean that you can’t add plenty of features to it. It’s all about using the available space properly. You get a total of eight SATA3 ports that are powered by the Intel C612 chipset and on top of that you get another eight SAS3 12Gbps ports thanks to the added LSI 3008 RAID controller. If that isn’t enough for you already, then you have plenty of possibilities to expand that with two PCIe Gen3 x16 and one PCIe Gen3 x8 slot.

The networking is a little bit special on this board as it comes with dual 10 Gigabit Ethernet thanks to the Intel X540 network controller. These ports are still RJ45 and backward compatible with normal RJ45 Gigabit Ethernet components. The ASRock Rack EPC612D4U-2T8R also features a dedicated IPMI 2.0 connection based on a Realtek RTL8211E controller that allows you to manage all your servers from a single connection, locally or remotely.

Four DDR4 DIMM slots with quad-channel support allows you to use this motherboard for memory-hungry applications and while the QVR list only has up to 16GB modules so far, you should also be able to use 32GB modules in this motherboard and get up to 128GB memory to support Intel’s latest Xeon E5 processors.

Overall, with the high performance processor, storage, and network all together in a mATX motherboard, the EPC612D4U-2T8R is perfect for the compact storage solution such as mid-range servers, web 2.0 storage servers, and other warm data servers. It also fits snugly into ASRock Rack’s previous released storage solution 1U12L-C612D4U, and currently has EPC612D4U, EPC612D4U-2T, EPC612D4U-8R and EPC612D4U-2T8R for customers’ selection.

Specifications

The specifications are taken directly from the manufacturers homepage and are as such subject to change in possible future revisions.

Key Features

There is always something that sets each and every motherboard apart for the rest, sometimes it’s in features and other times it is the software. On this motherboard, it is most likely the Network and storage capabilities as well as the form factor.

  • UATX 9.6”x9.6”
  • Single Socket LGA 2011 R3 Intel Xeon processor E5-1600/2600 v3 series
  • Supports Quad channel DDR4 2133/1866 RDIMM/LR DIMM/NV DIMM, 4 slots
  • Support 8 SATA3 by C612 and by 8 SAS3 LSI 3008
  • Supports 2 x PCIe 3.0 x16, 1 x PCIe 3.0 x 8
  • ASPEED 2400. Integrated IPMI 2.0 with KVM and Dedicated LAN (RTL8211E)
  • Supports Intel Dual 10GLAN by Intel X540

Packaging and Accessories

A server motherboard doesn’t need to come in a fancy box with flashy images and slogans, and that is why we don’t get that here. The EPC612D4U-2T8R comes in a simple black box with a little graphic and basic keywords.

The rear of the box doesn’t provide much more information either, but it will tell you a little bit about ASRock Rack as a company.

I really like that ASRock Rack also packed the motherboard properly like seen below. It provides that extra bit of protection during the transport from factory to end-customer over a motherboard that’s just loosely packed into a cardboard box.

Besides the motherboard itself, you’ll find an IO Shield, driver disk, and manual along with six SATA3 cables inside the box. It is nice to see a printed manual come along as it will help a lot with the basic task of finding the headers and pins you need to set it up.

Intel and Micron Introduce the Next Generation of Memory

Intel has teamed up with Micron to create the next generation of memory and today’s press event was to announce this to the world. It was done with big words and as we’ve learned from the recent Intel 750 SSD launch, they mean business when they say so.

Where the 750 SSD took NAND and moved it onto a better platform that could be better utilized by the CPU instead of being bottlenecked by SATA and SAS bus’, this introduction is something completely new. It is dubbed the 3D XPoint (3D Cross Point) and is truly the next generation memory.

It’s also about time that we get a new type of memory as the current NAND technology, while improved upon over time, already is over 25 years old.

3D XPoint is a new class of non-volatile memory that can provide speeds up to 1,000 times faster than current NAND technology. Not only is it faster, it is also a lot more durable and doesn’t have the trouble with a lot of writes as NAND does. This will bring game-changing performance to the market.

Not only is 3D XPoint said to be 1000 times faster and 1000 times more durable, it also has 10 times or more density than conventional memory.

The idea in itself isn’t new, but most people didn’t think it was possible. Intel and Micron had to come up with completely new materials and combinations as well as methods to combine them into a working product that could be mass produced.

Where normal memory just changes a part of the material used to indicate its state, 3D XPoint memory uses bulk material property change where the whole part changes instead of just being electron-based. This allows for more capacity in denser storage and the current production is spitting out 128Gbit sizes.

The 3D stacking is different from 3D NAND as it truly allows expansion in all direction without any effect to the performance. The unique switches inside are the key for this and it’s where the strength comes from. 3D NAND allows for more capacity, but not more speed, and that is why we need this new technology.

The pure nature of the technology also allows for much better data security as nothing will be lost in case of power failures. It can be used for both storage and system memory and as such could be the next big thing.

This isn’t just a proof of concept or a fancy powerpoint presentation with an idea, these are actual memory chips that currently are being produced in the joint factory of Micron and Intel. Both companies will release products based on this new technology in 2016 and they don’t expect any shortages in supply. That’s great news.

The final thing you might be asking yourself, what is the price and how does it really place itself in usability in comparison with other memory types. Both of these questions can be answered in one, it places itself between DRAM and NAND, so that’s not so bad news.

Supermicro X10SDV-F Xeon D-1540 SoC Motherboard Review

Introduction, Specifications, and Packaging


The motherboard I’m taking a look at today is a little something special. It is a tiny board yet it packs a punch that is worthy of a large right hook. The Supermicro X10SDV-F is a Mini-ITX motherboard with an onboard Intel Xeon D-1540 System-on-a-Chip (SoC) processor and lot of other goodies to go along with it. The Intel Xeon D-1540 is an octa-core processor with sixteen threads. It runs at 2GHz with a boost of up to 2.6.GHz. You will also need a lot of RAM to go along with that kind of CPU and that isn’t any problem for this tiny Supermicro board. It supports up to 128GB DDR4 memory with its 4 DIMM slots. You also get six high-speed SATA3 ports, a PCIe 3.0 x4 M.2 slot, a full speed PCIe Gen3 x16 slot and two Intel Gigabit Ethernet LAN ports. The bigger brother of this board is called the X10SDV-TLN4F which also features two integrated 10 Gigabit Intel Ethernet ports for those that need even more connectivity from a tiny board. Further, you get two USB 3.0 connectors on the rear I/O while USB 2.0 is provided via two headers. The remote management with a dedicated Realtek Ethernet port and built-in Aspeed graphics solution for maintenance and setup is another great feature and one that shouldn’t be missing on a server-grade motherboard.

There isn’t much you won’t find here. You can also choose whether you want to power it via a standard 24-pin power header or rather use 12V connectors depending on your power supply options. With a thermal design of just 45 watts TDP, the CPU can be passively cooled, but it is recommended that you cool the chassis properly.

Specifications

The specifications are taken directly from the manufacturers homepage and are as such subject to change in possible future revisions.

Key Features

There is always something that sets each and every motherboard apart for the rest, sometimes it’s in features and other times it is the software. On this motherboard, it is most likely the SoC and up to 128GB DDR4 memory support in this form factor.

  • mITX
  • Intel Xeon D-1540 SoC with 45W TDP
  • Up to 128GB DDR4 ECC Memory
  • ATX and 12V Dual power abilities and low power consumption
  • Dedicated IPMI Remote Management and SuperDoctor 5
  • 6 SATA3 ports, M.2 slot, USB 3.0 on-board, Dual Intel i350-AM2 Gigabit LAN
  • PCI-E gen3 x16 slot

Packaging and Accessories

As reviewers we don’t always get the retail package on these products as we simply don’t need them and the companies wish to get the products to us as fast as possible for review. In this case, I got the motherboard in a simple OEM box without any accessories. Normally however the motherboard includes two 57.5 cm flat SATA cables and a standard I/O shield.

Gigabyte MU70-SU0 (Intel C612) Server Motherboard Review

Introduction, Specifications, and Packaging


Intel’s C612 chipset brought along a lot of improvements for the workstation and server platform, not just support for Haswell-EP Xeon E5 processors, but also DDR4 memory support at full speed. Today I’m taking a look at one of Gigabyte’s many great server solutions and while most people aren’t that familiar with Gigabyte’s server-side of things, I can assure you that it is a quite impressive line-up. This is the first time we’ve taken a look at a server motherboard, and while we’ve analysed workstation type boards, you will see us opening up our site to a lot more content revolving around servers, their components and everything in between. With that in mind, I’m pleased to start our new server motherboard review section off with a feature packed, small but mighty motherboard from Gigabyte; the MU70-SU0.

Specifications

The specifications are taken directly from the manufacturers homepage and are as such subject to change in possible future revisions.

Key Features

There is always something that sets each and every motherboard apart for the rest, sometimes it’s in features and other times it is the software. On this motherboard, it is most likely the 12 RAM slots that you can run at full speed at the same time.

  • Intel Xeon processor E5-1600 V3 & E5-2600 V3 product families
  • 12 x RDIMM/LRDIMM ECC DDR4 DIMM slots
  • 4 x GbE LAN port (Intel I210)
  • 9 x SATA III 6Gb/s
  • 1 x mSATA slot
  • 1 x USB 3.0 + 1 x USB 2.0 headers
  • 1 x internal USB 3.0 port
  • Aspeed AST2400 remote management controller

Packaging and Accessories

The box follows a very simple design and is perfect for the market it’s aimed at. There is no need for fancy artwork like we’re used to on consumer motherboards.

The rear of the box gives a little detail about 3 key points that are important to a board of this caliber including reliability, availability, and serviceability. These are all major factors for any IT admin needing a reliable product with quality components that is easy to set up and maintain.

Crucial and Lexar Showcase Latest Storage Products @ Computex 2015

Computex 2015 – While, at the Crucial booth, we were contemplating how far storage technology has come in just a few years. With the humble hard drive being the ‘standard’ storage device with its technology dating back to the mid 20th century. In more recent years, we have seen the compression in terms of size and explosion in storage capacities. Crucial showcase today what they have made a name for themselves in, reliable and affordable NAND storage solutions. We see a wide range of storage products on display; including conventional MX200 SSD’s and LEXAR USB flash drives and SD cards.

One thing that plagues memory manufacturers is price, but we have seen a sharp decline in the overall price of NAND technology that consumer SSD options have surpassed the £1 per 1GB mark and are almost to the same price points as conventional hard drives.

 

We look forward to reviewing more of Crucial’s products in the future. We will keep you updated with any news and events from the rest of Computex 2015.

Crucial Showcase its Latest Memory Products @ Computex 2015

Computex 2015 – Memory is one of the few technology sectors that take a massive leap every few years; once a new standard is released, it can take a few months or even years for manufacturers to reap the full potential out of a memory product line. Take DDR3, for example, that was released with speeds up to 1600MHz, then 1866MHz was released and lately 2133MHz has become the norm with DDR4 taking off from there. With the recent announcement that that new Intel Skylake enthusiast processor can support DDR4 memory, we are likely to see the DDR4 standard explored in a much greater way in the upcoming months.

A quick look after the Crucial stand and you are bombarded with quick snippets of information regarding memory and how it has progressed over the years.

We look forward to bringing you a review on the Crucial Ballistix Tracer memory as soon as possible after Computex. We will keep you updated with any news and events from the rest of Computex.

Samsung Has Quite the Lead in the World of SSDs

Samsung isn’t always the first company that comes to mind when thinking of SSDs, but they are doing more than great in this area. The market research firm IHS recently released the report for the global SSD market in 2014 where Samsung takes a strong lead over the competition.

There isn’t much doubt that this was started with the 840 EVO series that brought a high-capacity drive with good performance and an unbeatable price for the time. Samsung hasn’t been sitting still since then and launched several successful drives and the new NVMe-based SM951 SSD will no doubt help to continue that trend and strengthen Samsung’s foothold.

Samsung claimed the first place with a 34% market share and a full year revenue of $ 3,996,000,000, that is about the same as Intel and Sandisk put together who come in on a second and third place respectively with around $1.9 billion revenue and a market share of 17%.

Micron comes in on a fourth place with a 8% market share, Toshiba grabs the fifth with 7% and Lite-On comes in on a sixth place with just 6% market share. What surprised me most was Kingston’s place 8th place and just a market share of 3%, they have a lot of amazing drives on the market.

IHS’s future predictions don’t look like they change a lot and that Samsung will continue down the road to even more market share. The total SSD market revenue of 2014 was an impressive 11.5 billion and it is expected to raise close to 2 billion this year.