Subject: Memory | February 4, 2017 - 01:42 AM | Tim Verry
Tagged: XPoint, server, Optane, Intel Optane, Intel, big data
Last week Hexus reported that Intel has begun shipping Optane memory modules to its partners for testing. This year should see the launch of both these enterprise products designed for servers as well as tiny application accelerator M.2 solid state drives based on the Intel and Micron joint 3D memory venture. The modules that Intel is shipping are the former type of Optane memory and will be able to replace DDR4 DIMMs (RAM) with a memory solution that is not as fast but is cheaper and has much larger storage capacities. The Optane modules are designed to slot into DDR4 type memory slots on server boards. The benefit for such a product lies in big data and scientific workloads where massive datasets will be able to be held in primary memory and the processor(s) will be able to access the data sets at much lower latencies than if it had to reach out to mass storage on spinning rust or even SAS or PCI-E solid state drives. Being able to hold all the data being worked on in one pool of memory will be cheaper with Optane as well as it is allegedly priced closer to NAND than RAM and the cost of RAM adds up extremely quickly when you need many terabytes of it (or more!). Various technologies attempting to bring higher capacity non volatile and/or flash-based storage in memory module form have been theorized or in the works in various forms for years now, but it appears that Intel will be the first ones to roll out actual products.
It will likely be years before the technology trickles down to consumer desktops and notebooks, so slapping what would effectively be a cheap RAM disk into your PC is still a ways out. Consumers will get a small taste of the Optane memory in the form of tiny storage drives that were rumored for a first quarter 2017 release following its Kaby Lake Z270 motherboards. Previous leaks suggest that the Intel Optane Memory 8000P would come in 16 GB and 32 GB capacities in a M.2 form factor. With a single 128-bit (16 GB) die Intel is able to hit speeds that current NAND flash based SSDs can only hit with multiple dies. Specifically the 16GB Optane application accelerator drive is allegedly capable of 285,000 random 4K IOPS, 70,000 random write 4K IOPS, Sequential 128K reads of 1400 MB/s, and sequential 128K writes of 300 MB/s. The 32GB Optane drive is a bit faster at 300,000 4K IOPS, 120,000 4K IOPS, 1600 MB/s, and 500 MB/s respectively.
Unfortunately, I do not have any numbers on how fast the Optane memory that will slot into the DDR4 slots will be, but seeing as two dies already max out the x2 PCI-E link they use in the M.2 Optane SSD, a dual sided memory module packed with rows of Optane dies on the significantly wider memory bus is very promising. It should lie somewhere closer to (but slower than) DDR4 but much faster than NAND flash while still being non volatile (it doesn't need constant power to retain the data).
I am interested to see what the final numbers are for Intel's Optane RAM and Optane storage drives. The company has certainly dialed down the hype for the technology as it approached fruition though that may be more to do with what they are able to do right now versus what the 3D XPoint memory technology itself is potentially capable of enabling. I look forward to what it will enable in the HPC market and eventually what will be possible for the desktop and gaming markets.
What are your thoughts on Intel and Micron's 3D XPoint memory and Intel's Optane implementation (Micron's implementation is QuantX)?
- IDF 2016: Intel To Demo Optane XPoint, Announces Optane Testbed for Enterprise Customers
- Intel Optane (XPoint) First Gen Product Specifications Leaked
- Intel Z270 Express and H270 Express Chipsets Support Kaby Lake, More PCI-E 3.0 Lanes
Subject: General Tech | December 7, 2016 - 06:25 PM | Jeremy Hellstrom
Tagged: qualcomm, centriq, centriq 2400, server
The days when AMD and Intel were the two choices to build a server with are long gone. The ARM architecture has been making serious inroads as various vendors have begun to offer various solutions utilizing ARM designs, up to and including AMD for that matter. Today, Qualcomm have joined these ranks, announcing their first processor family designed to power a server. The Centriq 2400 series is based on a 10nm process node, with up to 48 cores. As The Inquirer points out, this is a rather impressive shot across Intel's bow as Qualcomm will ship a 10nm FinFET before Intel does.
"The Qualcomm Centriq 2400 series, the first in the Centriq product family that Qualcomm has been working on for four years, has up to 48 ARMv8-compliant cores targeting compute-intensive data centre applications that require power efficiency and is built on the 10nm FinFET manufacturing processor."
Here is some more Tech News from around the web:
- Broadcom quietly dismantles its 'Vulcan' ARM server chip project @ The Register
- Sony kills off secret backdoor in 80 internet-connected CCTV models @ The Register
- Mikko Hypponen On The Death Of Antivirus @ Tech ARP
- Christmas 2016 Mega Worldwide Giveaway @ NikKTech
Clean Sheet and New Focus
It is no secret that AMD has been struggling for some time. The company has had success through the years, but it seems that the last decade has been somewhat bleak in terms of competitive advantages. The company has certainly made an impact in throughout the decades with their 486 products, K6, the original Athlon, and the industry changing Athlon 64. Since that time we have had a couple of bright spots with the Phenom II being far more competitive than expected, and the introduction of very solid graphics performance in their APUs.
Sadly for AMD their investment in the “Bulldozer” architecture was misplaced for where the industry was heading. While we certainly see far more software support for multi-threaded CPUs, IPC is still extremely important for most workloads. The original Bulldozer was somewhat rushed to market and was not fully optimized, while the “Piledriver” based Vishera products fixed many of these issues we have not seen the non-APU products updated to the latest Steamroller and Excavator architectures. The non-APU desktop market has been served for the past four years with 32nm PD-SOI based parts that utilize a rebranded chipset base that has not changed since 2010.
Four years ago AMD decided to change course entirely with their desktop and server CPUs. Instead of evolving the “Bulldozer” style architecture featuring CMT (Core Multi-Threading) they were going to do a clean sheet design that focused on efficiency, IPC, and scalability. While Bulldozer certainly could scale the thread count fairly effectively, the overall performance targets and clockspeeds needed to compete with Intel were just not feasible considering the challenges of process technology. AMD brought back Jim Keller to lead this effort, an industry veteran with a huge amount of experience across multiple architectures. Zen was born.
Hot Chips 28
This year’s Hot Chips is the first deep dive that we have received about the features of the Zen architecture. Mike Clark is taking us through all of the changes and advances that we can expect with the upcoming Zen products.
Zen is a clean sheet design that borrows very little from previous architectures. This is not to say that concepts that worked well in previous architectures were not revisited and optimized, but the overall floorplan has changed dramatically from what we have seen in the past. AMD did not stand still with their Bulldozer products, and the latest Excavator core does improve upon the power consumption and performance of the original. This evolution was simply not enough considering market pressures and Intel’s steady improvement of their core architecture year upon year. Zen was designed to significantly improve IPC and AMD claims that this product has a whopping 40% increase in IPC (instructions per clock) from the latest Excavator core.
AMD also has focused on scaling the Zen architecture from low power envelopes up to server level TDPs. The company looks to have pushed down the top end power envelope of Zen from the 125+ watts of Bulldozer/Vishera into the more acceptable 95 to 100 watt range. This also has allowed them to scale Zen down to the 15 to 25 watt TDP levels without sacrificing performance or overall efficiency. Most architectures have sweet spots where they tend to perform best. Vishera for example could scale nicely from 95 to 220 watts, but the design did not translate well into sub-65 watt envelopes. Excavator based “Carrizo” products on the other hand could scale from 15 watts to 65 watts without real problems, but became terribly inefficient above 65 watts with increased clockspeeds. Zen looks to address these differences by being able to scale from sub-25 watt TDPs up to 95 or 100. In theory this should allow AMD to simplify their product stack by offering a common architecture across multiple platforms.
Subject: Storage | August 11, 2016 - 02:59 PM | Allyn Malventano
Tagged: FMS, SYS-2028U-TN24R4T+, SYS-1028U-TN10RT+, supermicro, SSG-2028R-NR48N, server, NVMe, FMS 2016
Supermicro was at FMS 2016, showing off some of their NVMe chassis:
The first model is the SYS-1028U-TN10RT+. This 1U chassis lets you hot swap 10 2.5" U.2 SSDs, connecting all lanes directly to the host CPUs.
Supermicro's custom PCB and interposer links all 40 PCIe lanes to the motherboard / CPUs.
Need more drives installed? Next up is the SYS-2028U-TN24R4T+, which uses a pair of PCIe switches to connect 24 U.2 SSDs to the same pair of CPUs.
Need EVEN MORE drives installed? The SSG-2028R-NR48N uses multiple switches to connect 48 U.2 SSDs in a single 2U chassis! While the switches will limit the ultimate sequential throughput of the whole package to PCIe 3.0 x40, we know that when it comes to spreading workloads across multiple SSDs, bandwidth bottlenecks are not the whole story, as latency is greatly reduced for a given workload. With a fast set of U.2 parts installed in this chassis, the raw IOPS performance would likely saturate all threads / cores of the installed Xeons before it saturated the PCIe bus!
More to follow as we wrap up FMS 2016!
Subject: General Tech | December 11, 2015 - 05:16 PM | Jeremy Hellstrom
Tagged: server, qualcomm
AMD and Intel have been fighting it out in the server room for a while and have had to shift their tactics towards more efficient processors which merely sip at power compared to the first decade of this century. Coming from the other direction IBM and ARM design teams have been increasing the power of their chips and their ability to work together to match AMD and Intel's performance while still trying to maintain a lead on power efficiency. Now, according to what DigiTimes has been hearing, Qualcomm is ready to take advantage of its ARM license to officially move into the server market. Their initial design will sport 24 cores, provide support for VM environments and will be Linux compatible. Keep an eye on Xilinx and Mellanox Technologies as they were the companies who have announced plans to release products based on Qualcomm's designs.
"Qualcomm, which announced plans to begin developing ARM-based chips for servers in November 2014, has started delivering server-use CPU samples to potential clients and has also set up a company in Guizhou, China to promote the CPUs exclusively."
Here is some more Tech News from around the web:
- NVIDIA Iray “Predictive Design” Demo On Shield Tablet @ Tech ARP
- All eyes on the jailbroken as iOS, Mac OS X threat level ratchets up @ The Register
- Steam Escrow System Drives Impatient Users To Fake Trading Sites Serving Malware @ Slashdot
- Adobe Flash Player update includes 79 fixes, for crying out loud @ The Inquirer
- Google cloaks Android in Red Screen of malware Dearth @ The Register
- Facebook plans to open source its next-gen machine learning system @ The Inquirer
Introduction: This Is Not a NAS
The new WSS NAS series from Thecus contains some very interesting devices, and particularly so at the entry-level price with the unit we’re looking at today. WSS is the abbreviation for Windows Storage Server (in this case it’s 2012 R2), and this provides a huge increase in functionality compared to a standard NAS, as you might imagine.
Need a server? Just add a keyboard, mouse, and monitor
It’s really quite remarkable what Thecus is doing in partnership with Microsoft here in terms of value, as this entry 2-bay unit costs just $350. While this may seem high for a dual-bay NAS, we really aren’t talking about a NAS at all with this - which will be readily apparent to the user upon first powering it up. We are talking about a full-scale server here, replete with Windows Server 2012 R2 Essentials goodness. Of course a savvy user could easily deploy a small server in a home or office, and there are many advantages to managed solutions beyond the simple NAS appliances. But the advantage of a NAS is just that: it is significantly less complex and accessible for a consumer. The W2000 presents a very interesting option due to one particular aspect of its own accessibility: price. At $350 you are getting a very compact server with internal hardware much more akin to a standard desktop than you might imagine, and it ships installed with Microsoft's Windows Storage Server 2012 R2 Essentials.
What is “Storage” Server Essentials?
Ok, so I was a little confused as to the specific difference with the Storage version of the Server OS, unless it was simply a licensing distinction. My research first brought me to this quote from Microsoft:
“Windows Storage Server 2012 R2 Essentials is based on Windows Server 2012 R2. In fact, when it comes to functionality, you get key some features that aren’t included in these first two editions.”
After looking through the available documentation it appears as though Storage Server Essentials is, essentially, just Server Essentials with the distinction of being licensed differently. Microsoft TechNet defines it further:
“A computer that runs Windows Storage Server is referred to as a storage appliance. Windows Storage Server is based on the Windows Server operating system, and it is specifically optimized for use with network-attached storage devices. Windows Storage Server offers you a platform to build storage appliances that are customized for your hardware.”
Subject: General Tech | February 3, 2015 - 05:37 PM | Jeremy Hellstrom
Tagged: microsoft, server, operating system, sccm
There will be no Server 2015 release but sometime later in the year a preview version will be released for those wishing to have a peek at the new OS. We will see an update to SCCM arrive at roughly the same time as Windows 10 is released which will add support for managing Win10 machines and images and will allow a lot of sysadmins to sleep easier at night. The expected new features for the new server OS include the Docker image file format and containerization allowing you to run multiple programs on the same machine which are completely separated from each other and will be new to the Windows environment. Check out a short list of other features and a link to a more indepth look at the new containerization features expected from the new server OS at The Register.
"While it's looking like the final version of Windows 10 for client PCs could ship before the end of the year, it seems data center admins needn't hold their breaths. Microsoft confirmed on Friday that the next version of Windows Server won't arrive until 2016."
Here is some more Tech News from around the web:
- Systemd Getting UEFI Boot Loader @ Slashdot
- Intel brings Broadwell to businesses with 5th-gen Core chips with vPro @ The Register
- Raspberry Pi 2 hands-on review video @ The Inquirer
- Trouble comes in threes: Yet ANOTHER Flash 0-day vuln patch looming @ The Register
- Outlook for iOS branded a 'security nightmare' @ The Inquirer
- Ugly, incomplete, buggy: Windows 10 faces a sprint to the finish @ The Register
- TP-Link AV500 2-Port Powerline Network Adapter Kit @ Kitguru
- Microsoft to invest in Android firmware upstart Cyanogen @ The Register
- AEROCOOL Pimp My Rig Competition @ Kitguru
Subject: General Tech | October 27, 2014 - 04:35 PM | Jeremy Hellstrom
Tagged: Haswell-EX, Haswell-EP4S, Intel, server, xeon, Broadwell-DE, Skylake
Intel's release schedules have been slowing down, unfortunately in a large part that is due to the fact that the only competition they face in certain market segments is themselves. For high end servers it looks like we won't see Haswell-EX or EP4S until the second half of next year and Skylake chips for entry level servers until after the third quarter. Intel does have to fight for their share of the SoC and low powered chips, DigiTimes reports the Broadwell-DE family and the C2750 and C2350 should be here in the second quarter which gives AMD and ARM a chance to gain market share against Intel's current offerings. Along with the arrival of the new chips we will also see older models from Itanium, Xeon, Xeon Phi and Atom be discontinued; some may be gone before the end of the year. You have already heard the bad news about Broadwell-E.
"Intel's next-generation server processors for 2015 including new Haswell-EX (Xeon E7 v3 series) and -EP4S (Xeon E5-4600 v3 series), are scheduled to be released in the second quarter of 2015, giving clients more time to transition to the new platform, according to industry sources."
Here is some more Tech News from around the web:
- iOS 8.1 @ The Inquirer
- How to Get Open Source Android @ Linux.com
- Mozilla to make Firefox OS a tasty filling for a Raspberry Pi @ The Inquirer
- Pesky POS poison won't Backoff @ The Register
- Cisco patches three-year-old remote code-execution hole @ The Register
- Netgear Nighthawk R7000 AC1900 @ Kitguru
- Tech ARP 2014 Mega Giveaway Contest
- WIN a 1TB monster Samsung EVO 840 SSD @ The Register
Subject: Processors | May 8, 2014 - 04:26 AM | Tim Verry
Tagged: TrustZone, server, seattle, PCI-E 3.0, opteron a1100, opteron, linux, Fedora, ddr4, ARMv8, arm, amd, 64-bit
AMD showed off its first ARM-based “Seattle” processor running on a reference platform motherboard at an event in San Francisco earlier this week. The new chip, which began sampling in March, is slated for general availability in Q4 2014. The “Seattle” processor will be officially labeled the AMD Opteron A1100.
During the press event, AMD demonstrated the Opteron A1100 running on a reference design motherboard (the Seattle Development Platform). The hardware was used to drive a LAMP software stack including an ARM optimized version of Linux based on RHEL, Apache 2.4.6, MySQL 5.5.35, and PHP 5.4.16. The server was then used to host a WordPress blog that included stream-able video.
Of course, the hardware itself is the new and interesting bit and thanks to the event we now have quite a few details to share.
The Opteron A1100 features eight ARM Cortex-A57 cores clocked at 2.0 GHz (or higher). AMD has further packed in an integrated memory controller, TrustZone encryption hardware, and floating point and NEON video acceleration hardware. Like a true SoC, the Opteron A1100 supports 8 lanes of PCI-E 3.0, eight SATA III 6Gbps ports, and two 10GbE network connections.
The Seattle processor has a total of 4MB of L2 cache (each pair of cores shares 1MB of L2) and 8MB L3 cache that all eight cores share. The integrated memory controller supports DDR3 and DDR4 memory in SO-DIMM, unbuffered DIMM, and registered ECC RDIMM forms (only one type per motherboard) enabling the ARM-based platform to be used in a wide range of server environments (enterprise, SMB, and home servers et al).
AMD has stated that the upcoming Opteron A1100 processor delivers between two and four times the performance of the existing Opteron X series (which uses four x86 Jaguar cores clocked at 1.9 GHz). The A1100 has a 25W TDP and is manufactured by Global Foundries. Despite the slight increase in TDP versus the Opteron X series (the Opteron X2150 is a 22W part), AMD claims the increased performance results in notable improvements in compute/watt performance.
AMD has engineered a reference motherboard though partners will also be able to provide customized solutions. The combination of reference motherboard and ARM-based Opteron A1100 is known at the Seattle Development Platform. This reference motherboard features four registered DDR3 DIMM slots for up to 128GB of memory, eight SATA 6Gbps ports, support for standard ATX power supplies, and multiple PCI-E connectors that can be configured to run as a single PCI-E 3.0 x8 slot or two PCI-E 3.0 x4 slots.
The Opteron A1100 is an interesting move from AMD that will target low power servers. the ARM-based server chip has an uphill battle in challenging x86-64 in this space, but the SoC does have several advantages in terms of compute performance per watt and overall cost. AMD has taken the SoC elements (integrated IO, memory, companion processor hardware) of the Opteron X series and its APUs in general, removed the graphics portion, and crammed in as many low power 64-bit ARM cores as possible. This configuration will have advantages over the Opteron X CPU+GPU APU when running applications that use multiple serial threads and can take advantage of large amounts of memory per node (up to 128GB). The A1100 should excel in serving up files and web pages or acting as a caching server where data can be held in memory for fast access.
I am looking forward to the launch as the 64-bit ARM architecture makes its first major inroads into the server market. The benchmarks, and ultimately software stack support, will determine how well it is received and if it ends up being a successful product for AMD, but at the very least it keeps Intel on its toes and offers up an alternative and competitive option.
Subject: Processors | January 23, 2014 - 04:12 AM | Tim Verry
Tagged: server, piledriver, opteron 6300, amd, 32nm
AMD has updated its Opteron 6300 series lineup with two new processors with lower TDPs. Previously code-named "Warsaw," the Opteron 6370P and Opteron 6338P boast 99W TDPs and 12 and 16 Piledriver cores respectively.
The chips are similar to the existing Opteron 6300-series chips including the 32nm manufacturing process, dual die design, and the use of AMD's older Piledriver CPU cores instead of the latest Steamroller cores found in AMD's new Kaveri APUs. According to Supermicro, the lower 99W TDP parts offer up to 27% higher performance/watt compared to the existing "Abu-Dhabi" 6300 CPUs.
The Opteron 6338P is a twelve core processor clocked at 2.3 GHz base and 2.8 GHz turbo. The Opteron 6370P is a sixteen core part clocked at 2.0 GHz base and 2.5 GHz turbo. As such, the chips are two six and two eight-core silicon dies in one package respectively. The chips have 16MB of L3 cache and support the same instruction sets as the existing 6300 lineup including FMA3, BMI, and F16c. The new chips use AMD's Socket G34 which supports up to 4 sockets (dual die processors) per motherboard.
The new 99W 12-core 6338P and 16-core 6370P are available now for $377 and $598 respectively. The chips will be used in servers from Supermicro and Sugon, and purchasable directly from system integrators including Avnet and Penguin. AMD is aiming these chips at large data centers and cloud computing tasks. While the drop to 99W from the top-end series' 140W TDP does not seem like much, it makes a dramatic difference in the data center world where the electricity costs for racks of servers adds up rapidly.