Product Overview
The Supermicro Superserver SYS-F647G2-FT+ FatTwin is a high-density 2 hot-plug system node configuration with up to 1TB ECC DDR3 memory, 6x 2.5" Hot-swap SATA HDDs Bays, 2000W Redundant Platinum Level (94%+) Power Supplies and much more.
The Supermicro Superserver SYS-F647G2-FT+ FatTwin is a high-density 2 hot-plug system node configuration with up to 1TB ECC DDR3 memory, 6x 2.5" Hot-swap SATA HDDs Bays, 2000W Redundant Platinum Level (94%+) Power Supplies and much more.
Its high efficiency, highly effective shared cooling architecture allows the FatTwin to operate in high-temperature, free-air cooled environments up to 47°C, providing considerable cost savings and improved TCO. Versatile configurations allow the FatTwin to be optimized for many different environments including Data Center, Cloud Computing, and HPC; Engineering, Research, and GPU projects; File and Storage Server deployments; and General Server and Enterprise Server applications.
Key Features
2 Hot-plug System Nodes in 4U Front I/O. 6x GPU/Xeon Phi per node. Each node supports: Dual socket R (LGA 2011) supports Intel® Xeon® processor E5-2600 and E5-2600 v2 family Up to 1TB ECC DDR3, up to 1866MHz; 16x DIMM sockets 6x PCI-E 3.0 x16 for Double-width GPU/Xeon Phi; 2x PCI-E 3.0 x8 slots Front I/O ports: Dual GbE LAN, 2x USB 2.0, and 1x VGA connector Built-in Server management tool (IPMI 2.0, KVM/media over LAN) with dedicated LAN port 6x 2.5" Hot-swap SATA HDDs Bays 2000W Redundant Power Supplies Platinum Level (94%+)
-
BIOS Features
ACPI 1.0
ACPI 2.0
APM 1.2
PCI 2.2
Plug and Play (PnP)
SMBIOS 2.3
UEFI
USB keyboard support
-
BIOS Type
128MB SPI Flash with AMI UEFI
-
Buttons
Power On/Off button
UID button
-
Certifications
80 Plus Platinum (94%+)
-
Chipset
Intel C602
-
Colour
Black
-
Coprocessor Series
Intel Xeon Phi
-
CPU Family
Intel Xeon E5
-
CPU Manufacturer
Intel
-
CPU Quantity (Maximum)
2
-
CPU Series
Intel Xeon E5-2600
Intel Xeon E5-2600 v2
-
CPU Socket
Dual Socket R (LGA 2011)
-
Drive Bays
6 x 2.5" Hot-swap SATA
-
Error Detection
Corrects single-bit errors
Detects double-bit errors (using ECC memory)
-
Expansion Slots
2 x (x8) PCI-E 3.0 FH, HL slots
6 x (x16) PCI-E 3.0 FH, FL slots (support 6x Double-width GPU/Xeon Phi)
-
Form Factor
4U
-
GPU Family
Matrox G-Series
NVIDIA Tesla Series
-
GPU Manufacturer
Matrox
NVIDIA
-
GPU Model
Matrox G200eW
Tesla K1 (Kepler)
Tesla K2 (Kepler)
Tesla K10 (Kepler)
Tesla K20 (Kepler)
Tesla K20X (Kepler)
Tesla M2075
Tesla M2090
-
GPU Quantity
6 x Double-width GPU/Xeon Phi per node
-
Input/Output
1 x Fast UART 16550 serial port
1 x RJ45 Dedicated IPMI LAN port
1 x VGA port
2 x RJ45 Gigabit Ethernet LAN ports
2 x SATA 3.0 ports (6Gbps)
2 x USB 2.0 ports
4 x SATA 2.0 ports (3Gbps)
-
IPMI
IPMI 2.0 with virtual media over LAN and KVM-over-LAN support
Nuvoton WPCM450 BMC
Support for Intelligent Platform Management Interface v.2.0
-
LED indicators
Power fail LED
Power status LED
System Overheat LED
UID LED
-
Manufacturer
Supermicro
-
Memory (Maximum)
1TB
-
Memory Sizes
1GB
2GB
4GB
8GB
16GB
32GB
64GB
-
Memory Slots
16
-
Memory Speed
800MHz
1066MHz
1333MHz
1600MHz
1866MHz
-
Memory Type
DDR3 ECC LRDIMM
-
Memory Voltage
1.5V
1.35V
-
Network Adapter
Intel i350 Dual Port Gigabit Ethernet
Realtek RTL8201F PHY (dedicated IPMI)
Supports 10BASE-T, 100BASE-TX, and 1000BASE-T, RJ45 output
Virtual Machine Device Queues reduce I/O overhead
-
Number of Nodes
2
-
Product Range (sub)
FatTwin
-
PSU Output Capacity
2000 Watts
-
RAID Support (SATA)
RAID 0 (AHCI)
RAID 1 (AHCI)
RAID 5 (AHCI)
RAID 10 (AHCI)
-
Redundant Power Supply?
Yes
-
System Bus
QPI up to 8 GT/s
-
System Cooling
8 x 8cm heavy duty fans with air shroud and PWM fan speed control