Subject: Motherboards | May 15, 2013 - 03:56 AM | Tim Verry
Tagged: server, open source hardware, open source, open compute project, open 3.0, amd
Throughout last year, AMD worked with the Open Compute Foundation to develop open source hardware for servers. The goal of the project was to bring lower-cost, efficient motherboards (compatible with AMD processors) to the server market. Even better, the AMD-compatible hardware is open source which gives companies and OEM/system integrators free reign to modify and build the hardware themselves. The latest iteration of the project is called Open 3.0 and motherboards based on the design(s) are available now from a number of AMD partners.
An AMD Open 3.0 motherboard.
According to a recent AMD press release, Open 3.0 motherboards will be available from AVnet.inc, Hyve, Penguin Computing, and Zt Systems beginning this week. The new motherboards strip out unnecessary and "over-provisioned" hardware to cut down on upfront hardware costs and electrical usage. Open 3.0 uses a base open source motherboard design that can then be further customized to work with a variety of workloads and in various rack/server configurations. Servers based on OPen 3.0 will range from 1U to 3U in size and can slot into standard 19" racks or Open Rack environments. The boards with their dual Opteron 6300-series processors will reportedly be suitable for High Performance Computing (HPC), Virtual Desktop Infrastructure (VDI), Cloud applications, and storage servers. AMD claims that its Open 3.0 motherboards can reduce the Total Cost of Ownership (TCO) of servers by up to 57% in data centers. AMD claims that a server based on Open 3.0 has a TCO of $4,589 while one based on a traditional OEM motherboard costs up to 57% more at $10,669. The AMD-provided example sound nice. Despite the example likely being the best-case-scenario, the idea behind the Open Compute Project and the AMD-specific Open 3.0 hardware does make sense. Customers should see more competition with motherboards that are cheaper to produce and run thanks to the open source nature. Further details on the status of Open 3.0 and the available hardware is being discussed at an invitation-only industry round-table this week between partners, interested enterprise customers, and a number of companies (including AMD, Broadcom, and Quanta).
For the uninitiated, the Open 3.0 hardware features a motherboard that measures 16" x 16.7" and is intended for 1U, 1.5U, 2U, and 3U servers. Each Open 3.0 board includes two AMD Opteron 6300 series processors, 24 DDR3 DIMM slots (12 per CPU, 4 channels with 3 DIMMs each), six SATA ports, 1 managed dual-channel Gigabit Ethernet NIC, up to four PCI-E slots, and a single Mezzanine connector for custom modules (eg. the Mellanox IO or Broadcom Management card). Board IO will include a single serial port and two USB ports.
I'm glad to see AMD's side of the Open Compute Project come to fruition with the company's Open 3.0 hardware. Anything to reduce power usage and hardware cost is welcome in the data center world, and it will be interesting to see what kind of impact the open source hardware will have, especially when it comes to custom designs from system integrators. Intel is also working towards open source server hardware along with Facebook and the Open Compute Project. It is refreshing to see open source gaining traction in this market segment, to say the least.
Subject: General Tech | April 10, 2013 - 04:14 PM | Tim Verry
Tagged: xeon-ex, xeon-ep, xeon, server, Intel, HPC, haswell
Intel officially announced its next-generation Xeon processors at IDF Beijing today. The new lineup includes the Haswell-based Xeon E3 1200 V3 family on the low end, and the Ivy Bridge-EP Xeon E5 and Ivy Bridge-EX Xeon E7 aimed at the mid-range general purpose and high-end HPC markets respectively. Intel did not disclose pricing or details on the new chips (such as core counts, cache, clockspeeds, number of SKUs etc.). However, the x86 chip giant did state that the new chips are coming later this year as well as teasing a few tidbits of information on the new Xeon chips.
The upcoming Xeon E3 processors will be part of the Xeon E3 1200 V3 family. These chips will be based on Haswell and are limited to one socket per board. Thanks to the Haswell architecture, Intel has managed to reduce power consumption by approximately 25% and increase video transcoding performance by about 25%. There will be at least one Xeon E3 1200 V3 series chip with a 13W TDP, for example.
Intel is also releasing a new media software development kit (SDK) for Linux and Windows machines that will provide a common platform for developers. It has allowed Intel to maximize the use of both the CPU and GPU for HD video transcoding as well as increasing the number of simultaneous video transcodes over previous generations. The new Xeon E3 1200 V3 (Haswell) chips will be available sometime before the end of 2013.
The next-generation Xeon E5 chips will be based on the 22nm Ivy Bridge-EP architecture. They will be positioned at general purpose computing in data centers (and possibly high-end workstations), and will be limited to 2 sockets per motherboard. The new Xeon E5 processors will incorporate Intel Secure Key and OS Guard technologies. OS Guard is the evolution of the company's existing Intel Execute Disable Bit security technology. Intel is also including AES-NI (AES-New Instructions), to improve the hardware acceleration of AES encrypt/decrypt operations. These mid-range Xeon chips will be available in Q3 2013.
Finally, the top-end Xeon E7 processors will be based on the 22nm Ivy Bridge-EX architecture. The upcoming processors are intended for high performance server and supercomputing applications where scalability and performance are important. The Ivy Bride-EX chips are compatible with motherboards that will have between 4 and 8 sockets and up to 12TB of RAM per node. Further, Intel has packed these processors with new RAS features, including Resilient System Technology and Resilient Memory Technology. The RAS features ensure stability and data integrity in calculations are maintained. Such features are important in scientific, real-time analytics, cloud computing, and banking applications, where performance and up-time are paramount and any errors could cost a company money. Intel has stated that the new Xeon E7 CPUs will be available in the fourth quarter of this year (Q4'13).
While I was hoping for more details as far as core count, clockspeeds, and pricing, the approximate release to market timeframe for the chips is known. Do you think you will be upgrading to the new Xeon chips later this year, or are your current processors fast enough for your server applications?
More information on the upcoming Xeon chips can be found in this Intel fact sheet (PDF).
Subject: Processors | November 6, 2012 - 01:15 PM | Tim Verry
Tagged: server, piledriver, opteron, datacenter, cpu, amd
AMD announced new server processors on Monday based on the same Piledriver architecture used in the Trinity APUs and Vishera desktop CPUs we recently reviewed. With the release of the Opteron 6300 series, AMD is bringing Piledriver to the server room.
The new chips – similar to the desktop counterparts – bring several performance improvements over the previous generation 6200 series Opterons based on the Bulldozer architecture. AMD is positioning the chips as a upgrade path to existing servers and on merits of performance-per-dollar efficiency. As is AMD's fashion, the new chips are competitively priced and "good enough" performance-wise. With 6300, AMD has stated the goal is to reduce the TCO, or Total Cost of Ownership for servers used in data centers, supercomputers, and enterprises by being compatible with existing AMD server platforms with a BIOS upgrade and representing efficiency improvements over previous chips.
The Opteron 6300 series CPUs themselves build upon the Vishera desktop parts by adding more cores and more L3 cache. The server parts will have up to 16 cores clocked at 2.8GHz base and 3.2GHz turbo. They will have TDP ratings between 85W and 140W and will feature prices from $500 to $1,400. On the cache front, the chips have a 16KB L1 data cache per core, 64KB L1 instruction cache per module, 1MB L2 cache per core, and a shared 16MB cache per socket. AMD has included a quad channel memory controller that supports DDR3 up to 1866 MHz and 1.5TB per server in 4P configurations. AMD has rounded out the chips with four x16 HyperTransport 3.0 links rated at 6.4 GT/s per link. Up to 4 processors per server will be supported, which means a maximum of 64 cores.
With Piledriver, AMD added a number of new instructions including FMA3, BMI, and F16c. The company has also implemented server tweaks to the Bulldozer design to improve branch prediction, instructions per clock, scheduling, and reduced the power draw at higher clockspeeds allowing for the chps to clock higher while staying within the same power envelope of the Bulldozer-based Opteron 6200 series.
AMD is using the same socket as the 6200 series processors, and the new chips can be deployed as an upgrade to the old servers without needing a new motherboard.
When pitting the new Opteron 6380 to the previous-generation 6278, AMD is claiming a number of performance increases, including a 24-percent and 40-percent improvement in SPECjob2005 and SPECpower_ssj2008 respectively.
Further, the company is claiming competitive performance in server workloads with the Intel competition. AMD offers up benchmarks showing the Opteron 6380 and Xeon E5-2690 trading wins, with the AMD part being slower in the STREAM benchmark, but being slightly faster in LAMPS and NAMD. The allure of the Opteron, according to AMD is that the AMD part is almost half the price of the Intel processor, and is hoping the lower priced parts will encourage adoption. AMD argues that the money saved could easily go towards more RAM or more storage (or simply be saved of course).
The company has announced that its first major design win is Big Red II supercomputer at Indiana University. Built by Cray, the Big Red II will feature 21,000+ Opteron 6300-series CPU cores paired with NVIDIA GPUs. It represents a massive increase in computing power over IU’s previous Big Red supercomputer with 4,100 CPU cores, and will be used for medical, physics, chemistry, and climate research. Beyond that, AMD has stated more that 30 hardware vendors are slated to introduce servers based on the new Piledriver-based Opteron processors including HP, Dell, Cray, SGI, Supermicro, Sugon, and (of course) SeaMicro. On the software side of things, AMD is working with Microsoft, VMware, Xen, Red Hat, and Openstack. The company also stated that it is leaning on the experience and knowledge gained from the HSA Foundation to improve software support and guide the future direction of Opteron development.
The Opteron 6300 series is an interesting release that brings several improvements to the company’s server chip offerings. At launch, there are 10 processors to choose from, ranging from the quad core 6308 clocked at 3.5GHz for $501 to the top-end 6386 SE with 16 cores (2.8GHz base, 3.5GHz max turbo) and a $1,392 price tag. The 6366HE is an interesting part as well. It is the same price as the 12-core, 115W TDP Opteron 6348, but its has 16 lower-clocked cores and an 85W TDP. With the non-HE edition processors with 16 cores starting at $703, the 6366HE for $575 is a decent deal if you need multi-threading more than a fewer number of higher clocked cores.
Another bit that I found intriguing is that in a few years, AMD will (likely, if all goes according to plan) be offering processors for just about every type of server. They will have low cost, low power ARM Cortex-A57 based chips, Accelerated Processing Units (APUs) well suited to mixed workloads including GPU-accelerated tasks, and CPU-only chips with lots of traditional x86-64 cores. It seems that Intel will continue to hold the high end on pure performance, but AMD and its SeaMicro server division have not given up competing in the server room by a long shot.
The Piledrive architecture and Vishera desktop CPU review and The future of AMD: Vishera and Beyond at PC Perspective.
Subject: General Tech | July 7, 2012 - 01:12 AM | Tim Verry
Tagged: windows server, windows pricing, windows, virtual machines, software, server, operating system, enterprise
Earlier this week we covered the pricing for Microsoft’s upcoming Windows 8 consumer-grade operating system. Now, the company has released pricing information for the enterprise side of things, mainly for its non-OEM SKUs of Windows Server 2012. With Server 2012, Microsoft has simplified its lineup with four versions – one of which is only for OEMs.
Live Migration will allow virtualized storage to be moved in and out of server instances in real time without restarts.
The three versions that businesses can purchase and install themselves includes Datacenter, Standard, and Essentials. The lowest-tier version is called Foundation and will the version that comes pre-installed from OEMs. The Datacenter version has the most features and is the most lenient on the licensing by allowing businesses the full Windows Server 2012 functionality as well as unlimited virtual server instances. You’ll have to pay for those features, however as the Datacenter SKU is priced at $4,809. On the low end is Essentials which strips out licensed use of virtual instances of Server 2012 and also limites the number of user accounts that can access the server to 25. It will cost $425, which isn’t terribly expensive but is obviously aimed at small businesses. Interestingly, Microsoft states that Essentials has a simplified interface that is “pre-configured” for running cloud services. In the middle of those two extremes is Windows Server 2012 Standard which will run $882 USD and allows two virtualized instances as well as the full Windows Server functionality.
While Microsoft has not released pricing for its OEM-only Foundation version, they have announced that it will be limited to a max of 15 user accounts and no virtualization rights. The table below details the above information in a simplified table, courtesy Microsoft.
|Edition||Feature Comparison||Licensing Model||Pricing (USD)|
|Datacenter||Unlimited virtual instances, full Windows functionality||Processor + CAL||$4,809|
|Standard||Two virtual instances, full Windows functionality||Processor + CAL||$882|
|Essentials||No virtualization rights, Simple interface pre-configured for cloud services||Server (25 user account limit)||$425|
|Foundation||No virtualization rights, general purpose server functionality||Server (15 user account limit)||Not Listed|
As Martin Brinkman explains, the top-two tiers are based on a processor licensing model which means that each version is allowed to run on up to two physical processors. The Datacenter version takes that a step further by allowing an unlimited number of virtual machines on those two physical processors while Standard allows two virtual machines on a system with up to two physical processors. To figure out how many licenses you will need to purchase, you can get by with half the number of physical processors. At around five Windows Server 2012 Standard licenses, it starts to become more economical to go with the Datacenter version if you will mostly be spinning up virtualized servers.
Interestingly, Windows Home Server is missing from the above list, and it looks like that is not a mistake. Microsoft has stated in its licensing FAQ (PDF) that it expects home and small business users to move to the Essentials ($425) version for their home server needs. Not exactly the answer that many users are going to want to hear. For those not wanting to spend that much, Microsoft is keeping Windows Home Server 2011 alive until the end of next year (12-31-13), and you will be able to buy Home Server 2011 in an OEM machine until 2025. Fortunately, a system builder version of Windows Home Server 2011 can be found for around $50 and it can support up to 10 users. On the other hand, it won’t have the neat Windows 8-based server features. It will be up to you to decide whether the $400+ price for Essentials is worth it for you home/small business needs.
Just as Microsoft has released a Consumer Preview version of Windows 8, you can download a Release Candidate of Windows Server 2012 to see what the new features are and if they are worth the money. More information on the pricing and various versions can be found here. What do you think of the new Windows Server SKUs?
Subject: Storage | July 3, 2012 - 12:21 AM | Tim Verry
Tagged: ssd, slc, server, sandisk, PCIe SSD, flash, enterprise, caching
Flash storage company Sandisk has recently jumped into the world of enterprise PCI-E caching SSDs – what they are calling Solid State Accelerators. Currently, they are offering a 200GB and 400GB model under the company’s Lightning PCIe series. The SSDs feature a proprietary Sandisk controller driving 24nm SLC NAND flash, a PCI-E 2.0 x4 interface, and maximum power draw of 15 watts.
The Lightning Accelerators use the NAND flash for Sandisk’s own foundry and offer a large performance boost for servers and workstations over hard drives and SATA SSDs. It is capable of 410 MB/s sequential reads or 110,000 IOPS. Further, when using 4KB and 8KB blocks, the drives can reach 23,000 and 17,000 read/write IOPS respectively. Other specifications include an average response time of 245 microseconds, and less than 30 millisecond maximum response times. The Solid State Accelerators also feature sustained read and write latencies as low as 50 microseconds.
Sandisk has built the drives so that they can be configured as boot drives, storage drives, or caching drives. The company supports up to 5 drives in a single system, for a maximum of 2TB of flash storage. In addition, Sandisk is offering up its Flashsoft software that allows the Lightning Accelerators to be used as caching drives on Windows-based systems. Unfortunately, that is an additional cost which is not included in the already pricey SSDs (good thing for corporate expense accounts!).
Speaking of pricing, the 200GB LP206M has an MSRP of $1,350 while the 400GB LP406M has an MSRP of $2,350. Both cards have five year warranties and a MTBF rating of 2 million hours. You can find more information on the Sandisk Website.
It will be interesting to see how this Sandisk accelerator stacks up to the likes of the Intel 910 and FusioIO drives! The FusionIO FX, for example, gives you 420GB of QDP MLC NAND for $2,495, which works out such that Sandisk has a slightly lower cost-per-gigabyte value and SLC flash. We will have to wait for some independant reviews to say which drive is actually faster, however.
Subject: Processors, Systems | May 29, 2012 - 05:15 PM | Ryan Shrout
Tagged: server, dell, copper, arm
Dell announced today that is going to help enable the world of the ARM-based server ecosystem by enabling key hyperscale customers to access and develop on Dell's own "Copper" ARM servers.
Dell today announced it is responding to the demands of our customers for continued innovation in support of hyperscale environments, and enabling the ecosystem for ARM-based servers. The ARM-based server market is approaching an inflection point, marked by increasing customer interest in testing and developing applications, and Dell believes now is the right time to help foster development and testing of operating systems and applications for ARM servers.
Dell is recognized as an industry leader in both the x86 architecture and the hyperscale server market segments. Dell began testing ARM server technology internally in 2010 in response to increasing customer demands for density and power efficiency, and worked closely with select Dell Data Center Solutions (DCS) hyperscale customers to understand their interest level and expectations for ARM-based servers. Today's announcement is a natural extension of Dell's server leadership and the company's continued focus on delivering next generation technology innovation.
While these servers are still not publicly available, Dell is fostering the development of software and verification processes by seeding these unique servers to a select few groups. PC Perspective is NOT one of them.
Each of these 3U rack mount machines includes 48 independent servers, each based around a 1.6 GHz quad-core Marvell Armada XP SoC. Each of the sleds (pictured below) holds four discrete server nodes, each capable of as much as 8GB of memory on a single DDR3 UDIMM. Each node can access one 2.5-in HDD bay and one Gigabit Ethernet connection.
Click for a larger view
Even though we are still very early into the life cycle of ARM architectures in the server room, Dell claims that these systems are built perfectly for web front-ends and Hadoop environments:
Customers have expressed great interest in understanding ARM-based server advantages and how they may apply to their hyperscale environments. Dell believes ARM infrastructures demonstrate promise for web front-end and Hadoop environments, where advantages in performance per dollar and performance per watt are critical. The ARM server ecosystem is still developing, and largely available in open-source, non-production versions, and the current focus is on supporting development of that ecosystem. Dell has designed its programs to support today's market realities by providing lightweight, high-performance seed units and easy remote access to development clusters.
There is little doubt that Intel will feel and address this competition in the coming years.
Subject: General Tech | January 3, 2012 - 01:58 PM | Jeremy Hellstrom
Tagged: Intel, server, cpu, Romley, Ivy Bridge-H2, xeon e3, xeon e5
The server room will be getting an update over the next year thanks to Intel releasing numerous CPU models based on different architectures. First up comes Romley with a total of seven 8-core Xeons, a half dozen 6-core Xeons including both the E5-1660 and 1650 as well as the E5-2640 and relatives, five 4-core Xeons and a single dual core CPU. That will take us until close to summer. By then Intel will be working on eleven different Ivy Bridge-H2 series CPUs including the Xeon E3-1290v2 as well as seven more higher end processors including Xeon E5-2470, which will take us towards the end of 2012.
In addition to the regular lineup, DigiTimes also lists four low power Xeons which will arrive in 2012 including the 8-core Xeon E5-2650L.
"Intel is set to launch 40 new processors including those for its upcoming Romley platform, in the first half of 2012 with the company to release 20 models each quarter, according to sources from server players."
Here is some more Tech News from around the web:
- UMC develops 0.11 micron aluminum process @ SemiAccurate
- Neural networks control a toy car @ Hack a Day
- Asustek to reduce motherboard production volume for 1Q12 @ DigiTimes
- Netgear Powerline AV 500 Adapter Kit Review @ TechReviewSource
- SteelSeries DESMO Gaming Eyewear Review @ Techgage
Subject: Systems | July 20, 2011 - 01:18 PM | Jeremy Hellstrom
Tagged: rack mount, server, 3U
If you read PC Perspective on a regular basis the chances are very good that you have purchased individual components and assembled yourself a PC from them. There is a lesser chance that you have built a server, especially one using a rack mountable casing. The terminology is different, less about ATX and more about rack unit or U. U is a measurement of size, with a 1U case bearing a remarkable resemblance to a pizza box, with a height of 1.75 inches (44.45 mm) with a width depending on the style of racks you are installing into, with 19" and 23" being the standard. OCC takes you through the assembly of both an ATX case as well as a 3U case and recommendations of OS and software based on the intended use of your new server in this article.
"Hardware is one of the most important items in building a server, regardless of its overall purpose. Without enough power or memory to run the necessary applications, the server would subsequently be useless. Depending on its intended usage, however, the hardware may vary slightly. It is often a good idea to sit down and create a plan beforehand. Personally, I live by a rule of thumb that purchasing more than needed is better than having the server become congested or even fail when it is needed the most. For this guide, I will be putting together two different types of servers. The first will be a gaming server, designed to host multiple instances of LAN-based games. The second will be a file/web server, intended to back up data and make items accessible from any computer in a network. As always, hardware can vary depending on your needs; this is just a general overview of a configuration that I would use, based on the available hardware that I have on hand. Keep in mind that any computer can function as a server as long as the necessary software is installed. Therefore, you may even have old parts that can be reused for your server."
Here are some more Systems articles from around the web:
- HP Pavilion Elite h8-1010 Review @ TechReviewSource
- Sapphire Edge-HD2 Mini PC Review @ HardwareHeaven
- AMD A8-3850, Sapphire A75, G.Skill Flare and 2600MHz+ DDR @ Tweaktown
- Sapphire Edge HD2 Mini PC Review @Hi Tech Legion
- Sapphire Edge HD2 Mini PC @ Pro-Clockers
- Project BMW MPower level 10 PC a Mod project @ XtremeComputing
Subject: Processors | May 25, 2011 - 03:28 PM | Jeremy Hellstrom
Tagged: xeon, server, xeon x7, x7 4870, Intel
AnandTech got their hands on four of the the brand new 32nm Intel Xeon X7 4870, 10 cores clocked to 2.4GHz; perhaps a delayed 'Tick" but a tick nonetheless. Not only did they test the new chips they also had a chance to test it with Load Reduced DIMMs (LR-DIMM) as opposed to the old Fully Buffered style (FB-DIMMs) we were used to in days gone by. That spells higher capacity which is good considering the testbed they used can support up to 2TB of RAM to keep the 4 CPUs fed. This is a high end server part, not really competeing against AMD as a similar Opteron system would cost about 1/2 as much with performance reduced about the same as well. Check out this beast, but keep in mind a single CPU will set you back more than you paid for your whole system.
"Only one year later, Intel is upgrading the top Xeon by introducing Westmere-EX. Shrinking Intel's largest Xeon to 32nm allows it to be clocked slightly higher, get two extra cores, and add 6MB L3 cache. At the same time the chip is quite a bit smaller, which makes it cheaper to produce. Unfortunately, the customer does not really benefit from that fact, as the top Xeon became more expensive. Anyway, the Nehalem-EX was a popular chip, so it is no surprise that the improved version has persuaded 19 vendors to produce 60 different designs, ranging from two up to 256 sockets."
Here are some more Processor articles from around the web:
- Intel Core i5-2390T @ iXBT Labs
- Inexpensive AMD Processor Roundup @ iXBT Labs
- AMD Phenom II X4 980 BE 3.70 GHz @ techPowerUp
- AMD Phenom II X2 560 Black Edition AM3 Processor Review @ eTeknix
- Workstation & Server CPU Comparison Guide @ TechARP+
- CPU Performance Comparison Guide @ TechARP
- Intel's Silvermont: A New Atom Architecture @ AnandTech