Supermicro 6027R-N3RF4+: Layout And Overview
The Supermicro 6027R-N3RF4+ is a complete barebones kit composed of an X9DRW-3LN4F+ motherboard and CSE-829BTQ-R920WB chassis. Both parts are available separately, but Supermicro sells the combination with cabling already run, routed, and secured to optimize airflow. This does save quite a bit of time compared to purchasing each piece and assembling. Supermicro also offers Intel CPUs and other components on its price list, so it is possible to not only purchase the company's components, but also to configure a fully-assembled system.
Let's take a look at the 6027R-N3RF4+ barebones package.
Looking at the front of the CSE-829BTQ-R920WB chassis, ten 3.5" disk trays are perhaps most prominent. Generally, this provides room to install two boot drives and a fully-populated eight-channel SAS controller. Three-and-a-half-inch disks are generally used as mass storage, with 10 000 and 15 000 RPM spindles becoming less common (in favor of 2.5" SAS drives or solid state drives when performance is the priority), making 3.5" the form factor of choice for larger, slower 7200 RPM SAS or SATA devices.
You can also see Supermicro's LCD status panel, along with USB ports, status LEDs, and buttons. The company makes room for a slim optical DVD drive near the top of the chassis, which is less common these days due to the remote image mounting capabilities of many management interfaces.
A quick look at the drive tray makes it pretty obvious that it'd be difficult to install a 2.5" device without utilizing an adapter. Supermicro does sell adapter hardware, but given the dominance of 2.5" storage for high-IOPS workloads, I do hope Supermicro's next design incorporates the ability to use 2.5" drives in 3.5" bays through either a unified drive sled or adapter.
We clearly see that the CSE-829BTQ-R920WB is built to accommodate custom form factor motherboards (WIO+). Proprietary form factors are very common when vendors design systems with expansion slots using riser cards, which the 6027R-N3RF4+ does utilize. This is important because the number and location of expansion slots via risers is not standardized, as you might expect on the desktop. When you purchase a system based on a proprietary form factor from a vendor like Supermicro, Dell, HP, or IBM, the ability to upgrade later on is often limited as a result.
There are four hot-swap 100 CFM 80 mm fans that dissipate heat from not just the motherboard, but also the passive CPU heat sinks through the use of clever ducting with an air shroud. Hot-swap fans in a chassis of this price are standard because, if a fan fails, you need the flexibility to pull it out and install a new one from the top of the chassis without touching the motherboard. This is doubly important in 1U and 2U enclosures, where CPUs are often passively cooled.
Moving to the rear of the chassis, the Supermicro 6027R-N3RF4+ includes redundant hot-swap 1U 920 W power supplies with 80 PLUS Platinum ratings, configured one on top of the other. These PSUs have a replacement part number of PWS-920P-SQ, with the "SQ" designating Supermicro products marketed as "Super Quiet". We find that interesting in a 2U server chassis destined for data centers. The 80 PLUS Platinum rating means that these PSUs can achieve up to 94% efficiency, an important metric in racks, where power delivery and cooling are finite. PSU efficiency plays an important role in lowering overall infrastructure load.
In order to remove one of the platform's power supplies, push the latch securing it and pull the unit's handle. Inserting a power supply is as simple as pushing it into a vacant slot.
I agree. Just reduce it a little bit but don't make it too hard to see
As an AMD fan, I wish we could. But while Magny-Cours was competitive with the last gen Xeons, AMD doesn't really have anything that stacks up against the E5. In pretty much every workload, E5 dominates the 62xx or the 61xx series by 30-50%. The E5 is even price competitive at this point.
We'll just have to see how Piledriver does.
Having said that I would suggest you include expected PPD for the given TPF since that is what folders look at when deciding on hardware. Or you could just devote 48 hours from each machine to generate actual results for F@H and donate those points to your F@H team (yes Tom's has a team and visibility is our biggest problem).