This document outlines the UEFI configuration menus and settings for the OnLogic Axial AX300 Series Edge Servers. The terms “BIOS” and “UEFI” are often used interchangeably in this document to describe the firmware interface used to start up a system. UEFI is a more advanced firmware interface that provides similar functionality to legacy BIOS firmware. UEFI offers a more flexible and feature-rich environment for controlling hardware and boot processes. It is designed to work with larger hard drives and newer hardware features, such as 64-bit processors, Secure Boot, and virtualization. The Axial AX300 Series Edge Servers utilize modern UEFI firmware for enhanced functionality and to support the most recent technologies.
To enter the UEFI Setup menu, press the or button on a keyboard while the system is powering on.
Once you enter the UEFI SETUP UTILITY, the Main screen will appear and display the system overview. The Main screen provides system overview information and allows you to set the system time and date.
Enter this item to view the motherboard information.
Enter this item to view the processor information.
Enter this item to view the memory information.
The Advanced configuration screen allows for configuration of the following:
CPU Configuration
Platform Power Configuration
DRAM Configuration
Chipset Configuration
Select the number of cores to enable in each processor package.
Intel Hyper Threading Technology allows multiple threads to run on each core, improving overall performance on threaded software.
Enables Intel Trusted Execution Technology Configuration.
Intel Virtualization Technology allows a platform to run multiple operating systems and applications in independent partitions, enabling one computer system to function as multiple virtual systems.
Use this item to enable Safer Mode Extensions.
Select the number of E-Cores to enable in each processor package.
Use this item to enable or disable Memory Encryption (TME).
Enable Enhanced Halt State (C1E) for lower power consumption.
Use this item to enable or disable Software Guard Extensions (SGX).
Enable C7 deep sleep state for lower power consumption.
Automatically prefetch data and code for the processor. Enable for better performance.
Automatically prefetch the subsequent cache line while retrieving the currently requested cache line. Enable for better performance.
Use this item to enable or disable AES-NI support.
Intel SpeedStep technology allows processors to switch between multiple frequencies and voltage points for better power saving and heat dissipation. CPU turbo ratio can be fixed when Intel SpeedStep Technology is disabled and Intel Turbo Boost Technology is enabled.
Intel Turbo Boost Technology enables the processor to run above its base operating frequency when the operating system requests the highest performance state.
Select this item to configure the AVX P1 level.
Select this item to configure the hardware-supported level.
Select this item to enable or disable Dynamic SST-PP.
Note: HWP Native Mode is a prerequisite for enabling Dynamic SST-PP.
Select this item to enable or disable SST-BF.
Note: HWP Native Mode is a prerequisite for enabling SST-BF; HWP Native Mode with No Legacy is a prerequisite for configuring SST-BF.
Select this item to enable or disable the BIOS from configuring SST-BF High Priority Cores so that software does not have to configure them.
This item supports the following selections:
Disable: Hardware chooses a P-state based on OS Request (Legacy P-States).
Native Mode: Hardware chooses a P-state based on OS guidance.
Out of Band Mode: Hardware autonomously chooses a P-state (no OS guidance).
Select this item to enable or disable the SST-CP feature.
Select this item to configure whether Monitor and MWAIT instructions automatically map to enable.
Select this item to configure the CPU C6 (ACPI C3) report to the OS.
This item specifies the Core C1E auto-promotion Control and whether it takes effect after reboot.
This item specifies the Package C State limit; the state Auto maps is program specific.
Select this item to enable or disable Thermal Monitor.
This allows users to decide which of the following controls EFB.
OS Controls EPB: Specifies IA32_ENERGY_PERF_BIAS is used.
BIOS Controls EPB: Specifies ENERGY_PERF_BIAS_CONFIG is used.
PECI Controls EPB: Specifies PCS53 is used.
This allows the user to use input from ENERGY_PERF_BIAS_CONFIG mode selection. Options include PERF/ Balanced, Perf/Balanced, or Power/Power.
Select this item to configure the Long Duration Power Limit. PL1 Power Limit is in Watts and the value may vary from 0 to Fused Value. If the value is 0, the fused value will be programmed. A value greater than the fused TDP value will not be programmed.
Select this item to configure the Long Duration Maintained value. PL1 value is in seconds. The value may vary from 0 to 448, indicating the time window over which the TDP value should be maintained.
Select this item to configure the Short Duration Power Limit. PL2 Power Limit in Watts. The value may vary from 0 to Fused Value. If the value is 0, BIOS programs 120% * TDP.
Select this item to configure the Short Duration Maintained value. PL2 value is in seconds. The value may vary from 0 to 0.438, indicating the time window over which the TDP value should be maintained.
Enable to enforce POR restrictions for DDR frequency and voltage programming.
If [Auto] is selected, the motherboard will detect the memory module(s) inserted and assign the appropriate frequency automatically.
Use this item to enable or disable Non Uniform Memory Access (NUMA).
Select 1LM or 2LM mode for Volatile memory. For 2LM memory mode, BIOS will attempt to configure 2LM, but if unsuccessful, volatile memory mode will fail back to 1LM.
Enable this item to allow interleaving to be performed at the highest possible MC * Channel interleaving ways.
Enable this item to allow 3-way channel interleaving. Alternatively, disable this item to default to the number of channels available per MC when DIMM interleaving is enabled.
This item allows you to select the Rank Interleaving setting.
Mirror Mode will set the entire 1LM/2LM memory in the system to be mirrored, consequently reducing the memory capacity by half. Disabled by default.
Enable or disable Memory Rank Sparing.
Patrol Scrub is a background activity initiated by the processor to seek out and fix memory errors.
Enable: Enables data scrambling for DDR4 and DDR5.
Disable: Disables this feature.
Auto: Sets it to the MRC default setting; current default is Enable.
Enable: Enables data scrambling for PMem.
Disable: Disables this feature.
Auto: Set it depending on stepping.
Use this item to select MMCFG Base. For AX300, the default of [Auto] is generally best for multi-CPU and 3+ GPU configurations to ensure sufficient resource availability in the OS.
Use this item to select MMCFG Size.
NOTE: To ensure sufficient resource for usage, it is recommended for users to disable the hot-plug option when setting this option to 128M. For AX300, the default of [Auto] is generally best for multi-CPU and 3+ GPU configurations to ensure sufficient resource availability in the OS.
Use this item to select MMIO High Base.
Use this item to select MMIO Granularity Size.
If the system has SR-IOV capable PCIE devices installed, this option enables or disables Single Root IO Virtualization Support.
If the system has resizable BAR capable devices installed, enable or disable this item to Re-Size BAR.
Use this to enable or disable the Onboard VGA function.
Use this to enable or disable the Onboard LAN function.
Intel Virtualization Technology for Directed I/O helps the virtual machine monitor better utilize hardware by improving application compatibility and reliability, and providing additional levels of manageability, security, isolation, and I/O performance.
Select this item to configure SATA or PCIE in the OCU port.
Select this item to configure PCIE Link Width.
PCIE0/1/2/3/4/5/6 Link Width: Select PCIe port Bifurcation for PCIE0/1/2/3/4/5/6.
MCIO1/2 Link Width: Select MCIO1/2 Link Width.
Select PCIE Link Speed.
OCU1/2/3 Link Speed: Configure PCIe Speed.
PCIE0/1/2/3/4/5/6 Link Speed: Select Link Speed for PCIE0/1/2/3/4/5/6.
M2_1 Link Speed: Select Link Speed for M2_1.
OCU1/2/3 Hot Plug: Enable or disable PCIE Hot Plug.
PCIE0/1/2/3/4/5/6, MCIO1-1/1-2/2-1/2-2 Hot Plug: Enable or disable PCIE and MCIO Hot Plug.
PCIE0/1/2/3/4/5/6, MCIO1-1/1-2/2-1/2-2 Surprise Hot Plug: Enable or disable PCIE and MCIO Surprise Hot Plug.
Select this item to configure the PCIE ASPM.
PCI-E ASPM Support (Global): Select this item to disable ASPM Support in all PCIe root ports.
OCU1/2/3 ASPM Support: Select this item to configure PCIE Active State Power Management settings.
PCIE0/1/2/3/4/5/6 ASPM Support: Enables or disables the ASPM support for all CPU downstream devices. Select [Auto] for the default value.
Enable or disable the onboard Dr. Debug LED.
Select this item to enable or disable the launch of the NVMe driver.
S.M.A.R.T stands for Self-Monitoring, Analysis, and Reporting Technology. It is a monitoring system for computer hard disk drives to detect and report on various indicators of reliability.
Use this item to enable or disable SATA Controllers.
Identify if the SATA port is connected to a Solid State Drive or Hard Disk Drive. Press <Ctrl+I> to enter RAID ROM during UEFI POST process.
Use this item to enable or disable Aggressive Link Power Management.
Select this item to configure the External SATA, Hot Plug, Spin Up Device, and SATA Device Type for the following:
OCU1 (SATA0_0/SATA0_1/SATA0_2/SATA0_3)
OCU2 (SATA0_4/SATA0_5/ SATA0_6/SATA0_7)
OCU3 (SATA1_4/SATA1_5/SATA1_6/SATA1_7)
SATA1_2
Allow the system to wake via a PCIE device and enable wake on LAN.
Note that AX300 on Windows Server 2022 does not support the ACPI sleep state, and therefore does not support WoL.
Use this item to enable or disable Ring-In signals to turn on the system from the power soft-off mode.
Use this item to enable or disable RTC (Real Time Clock) to power on the system.
Use this item to set parameters of Serial Port 1.
Serial Port
Use this item to enable or disable the serial port (COM).
Change Settings
Use this item to select an optimal setting for a Super IO device.
Use this item to set parameters of SOL.
SOL Port
Use this item to enable or disable the SOL port.
Change Settings
Use this item to select an optimal setting for a Super IO device.
Console Redirection
Use this option to enable or disable Console Redirection. If this item is set to [Enabled], you can select a COM Port to be used for Console Redirection.
Console Redirection Settings
Use this option to configure Console Redirection Settings, and specify how your computer and the host computer to which you are connected exchange information.
Terminal Type
Use this item to select the preferred terminal emulation type for out-of-band management. It is recommended to select [VT-UTF8].
Bits Per Second
Use this item to select the serial port transmission speed. The speed used in the host computer and the client computer must be the same. Long or noisy lines may require lower transmission speed. The options include [9600], [19200], [38400], [57600] and [115200].
Data Bits
Use this item to set the data transmission size. The options include [7] and [8] (Bits).
Parity
Use this item to select the parity bit. The options include [None], [Even], [Odd], [Mark] and [Space]. A parity bit can be sent with the data bits to detect some transmission errors. Mark and Space Parity do not allow for error detection; they can be used as an additional data bit.
Even: Parity bit is 0 if the number of 1s in the data bits is even.
Odd: Parity bit is 0 if the number of 1s in the data bits is odd.
Mark: Parity bit is always 1.
Stop Bits
This item indicates the end of a serial data packet. The standard setting is [1] Stop Bit. Select [2] Stop Bits for slower devices.
Flow Control
Use this item to set the flow control to prevent data loss from buffer overflow. When sending data, if the receiving buffers are full, a "stop" signal can be sent to stop the data flow. Once the buffers are empty, a "start" signal can be sent to restart the flow. Hardware flow uses two wires to send start/stop signals. The options include [None] and [Hardware RTS/CTS].
VT-UTF8 Combo Key Support
Use this item to enable or disable the VT-UTF8 Combo Key Support for ANSI/VT100 terminals.
Recorder Mode
Use this item to enable or disable Recorder Mode to capture terminal data and send it as text messages.
Resolution 100x31
Use this item to enable or disable extended terminal resolution support.
Putty Keypad
Use this item to select Function Key and Keypad on Putty.
Legacy Console Redirection
Use this option to configure Legacy Console Redirection Settings, and specify how your computer and the host computer to which you are connected exchange information.
Redirection COM Port
Select a COM port to display redirection of Legacy OS and Legacy OPROM Messages.
Resolution
On Legacy OS, the Number of Rows and Columns supported redirection.
Redirection After BIOS POST
If [LoadBooster] is selected, legacy console redirection is disabled before booting to the legacy OS. If [Always Enabled] is selected, legacy console redirection is enabled for the legacy OS. The default value is [Always Enabled].
Console Redirection
Use this option to enable or disable Console Redirection. If this item is set to [Enabled], you can select a COM Port to be used for Console Redirection.
Console Redirection Settings
Use this option to configure Console Redirection Settings, and specify how your computer and the host computer to which you are connected exchange information.
Out-of-Band Mgmt Port
Microsoft Windows Emergency Management Services (EMS) allows for remote management of a Windows Server OS through a serial port.
Terminal Type EMS
Use this item to select the preferred terminal emulation type for out-of-band management. It is recommended to select [VT-UTF8].
Bits Per Second EMS
Use this item to select the serial port transmission speed. The speed used in the host computer and the client computer must be the same. Long or noisy lines may require lower transmission speed. The options include [9600], [19200], [57600] and [115200].
Flow Control EMS
Use this item to set the flow control to prevent data loss from buffer overflow. When sending data, if the receiving buffers are full, a "stop" signal can be sent to stop the data flow. Once the buffers are empty, a "start" signal can be sent to restart the flow. Hardware flow uses two wires to send start/stop signals. The options include [None], [Hardware RTS/ CTS], and [Software Xon/Xoff].
Data Bits EMS
Statically set to 8.
Parity EMS
Statically set to None.
Stop Bits EMS
Statically set to 1.
Monitor the status of the hardware on your system, including the parameters of the CPU temperature, motherboard temperature, CPU fan speed, chassis fan speed, and the critical voltage.
Use this item to enable or disable the System Error feature. When it is set to [Enabled], it allows the user to configure Memory Error and PCIE Error log features.
Use this item to enable or disable Windows Hardware Error Architecture.
Use this item to enable or disable EMCA Logging.
Use this item to enable or disable IIO/PCH Error Support.
Use this item to enable or disable Memory Corrected Error.
Correctable Error Threshold (0 - 0x7FFF) used for sparing, tagging, and leaky bucket.
Use this item to enable or disable PCIe Correctable errors.
PCIE Correctable Error Threshold (0x01-0xFF) used for sparing, tagging, and leaky bucket.
Use this item to enable or disable PCIe Uncorrectable errors.
Use this item to enable or disable PCIe Fatal errors.
Use this item to enable or disable UEFI Network Stack.
Use this item to enable or disable IPv4 PXE boot support. If disabled, IPv4 PXE boot support will not be available.
Use this item to enable or disable IPv4 HTTP boot support. If disabled, IPv4 HTTP boot support will not be available.
Use this item to enable or disable IPv6 PXE boot support. If disabled, IPv6 PXE boot support will not be available.
Use this item to enable or disable IPv6 HTTP boot support. If disabled, IPv6 HTTP boot support will not be available.
Specifies the wait time and press the ESC key to abort the PXE boot.
Specifies the number of times the presence of physical storage devices are verified on a system reset or power cycle.
Use this item to enable or disable Intel Volume Management Device Technology in a specific Stack. When [Enabled], users are allowed to configure the options below.
Use this item to enable or disable Intel Volume Management Device Technology on a specific root port.
Use this item to enable or disable Hot Plug for specific Ports.
Use this item to enable or disable Intel Volume Management Device Technology for specific PCIe ports.
Provides health status for the drivers/controllers.
Press to configure Server CA.
Press to configure Client Cert.
Instant Flash is a UEFI flash utility embedded in Flash ROM. This utility enables UEFI firmware updates without entering operating systems. To flash UEFI firmware, download and save the new firmware version to bootable USB flash media. Then, using Instant Flash, you can update system firmware directly via Instant Flash menus through the UEFI Menu. Please note that the USB flash media must use a FAT32/16/12 file system. When the Instant Flash utility is executed, the utility will show the firmware files on the USB media and their respective information. Select the proper firmware file to update your UEFI, and reboot your system after the UEFI update process is completed.
Wait for BMC response for a specified timeout. BMC starts at the same time as BIOS during AC power ON. It takes approximately 90 seconds to initialize Host to BMC interfaces.
Use this item to enable or disable the FRB-2 timer (POST timer).
Enter a value between 1 to 30 minutes for FRB-2 Timer Expiration.
Use this item to configure how the system should respond if the FRB-2 Timer expires. This option is not available if the FRB-2 Timer is disabled.
If enabled, this starts a BIOS timer which can only be shut off by Management Software after the OS loads. It helps determine if the OS successfully loaded or follows the OS Boot Watchdog Timer policy.
Configure the OS Boot Watchdog Timer Expiration between 1 to 30 minutes. If the OS Boot Watchdog Timer is disabled, this item is not available.
Configure how the system should respond if the OS Boot Watchdog Timer expires. If the OS Boot Watchdog Timer is disabled, this item is not available.
Change this to enable or disable all features of System Event Logging during boot.
Use this to choose options for erasing SEL.
Use this to choose options for reactions to a full SEL.
Use this item to disable the logging of EFI Status Codes or log only error code or only progress or both.
Use this item to enable or disable PCIe Device Degrade Error Logging Support.
Select this item to enable or disable bonding. Please enable all LAN channels first when you want to enable bonding.
If [No] is selected, the IP address is assigned by DHCP. If you prefer using a static IP address, toggle to [Yes], and the changes take effect after the system reboots. The default value is [No].
Select to configure BMC network parameters statically or dynamically (by BIOS or BMC). Configuration options: [Static] and [DHCP].
Static
Manually enter the IP Address, Subnet Mask, and Gateway Address in the BIOS for BMC LAN channel configuration.
Note: The default login information for the IPMI web interface is: Username: admin Password: admin
DHCP
IP address, Subnet Mask, and Gateway Address are automatically assigned by the network's DHCP server.
Note: When [DHCP] or [Static] is selected, do NOT modify the BMC network settings on the IPMI web page.
VLAN
Enable/Disable Virtual Local Area Network. If [Enabled] is selected, configure the items below.
IPV6 Support
Enable/Disable LAN1 IPV6 Support.
Manual Setting IPMI LAN(IPV6)
Select to configure LAN channel parameters statically or dynamically (by BIOS or BMC). Unspecified options will not modify any BMC network parameters during the BIOS phase.
Select this KCS interface state after POST end. If [Enabled] is selected, the BMC will remain KCS interface after POST stage. If [Disabled] is selected, the BMC will disable KCS interface after POST stage.
This allows you to set the power state after an unexpected AC/power loss. If [Power Off] is selected, the AC/power remains off when the power recovers. If [Power On] is selected, the AC/power resumes and the system starts to boot up when the power recovers. If [Last State] is selected, it will recover to the state before AC/power loss.
Use this item to Load BMC Default Settings.
Set or change the password for the administrator account. Only the administrator has authority to change the settings in the UEFI Setup Utility. Leave it blank and press enter to remove the password.
Set or change the password for the user account. Users are unable to change the settings in the UEFI Setup Utility. Leave it blank and press enter to remove the password.
Use this item to enter the Secure Boot configuration page and enable or disable support for Secure Boot.
Enable to support Windows 8 or later versions Secure Boot.
Expert users can modify Secure Boot Policy variables without full authentication.
Factory Key Provision
Install factory default Secure Boot keys after the platform reset and while the System is in Setup mode.
Install Default Secure Boot Keys
Please install default secure boot keys if it’s the first time using secure boot.
Clear Secure Boot keys
Force System to Setup Mode - clear all Secure Boot Variables. Change takes effect after reboot.
Enroll Efi Image
Allow the image to run in Secure Boot mode. Enroll SHA256 Hash certificate of a PE image into Authorized Signature Database (db).
Export Secure Boot variables
Copy NVRAM content of Secure Boot variables to files in a root folder on a file system device.
This allows you to configure the Smbios Event Log Settings. When entering the item, you will see the following:
Use this item to enable or disable all features of the SMBIOS Event Logging during system boot.
The options include [No], [Yes, Next reset] and [Yes, Every reset]. If Yes is selected, all logged events will be erased.
Use this item to choose options for reactions to a full Smbios Event Log. The options include [Do Nothing] and [Erase Immediately].
Choose option to enable/disable logging of System boot event.
Press to view the Smbios Event Log records.
Display the available boot devices, configuration settings, and boot priority.
Use this item to set the system boot order.
Specifies the Boot Device Priority sequence from available UEFI Application.
Configure the number of seconds to wait for the UEFI setup utility.
If this item is set to [On], it will automatically activate the Numeric Lock function after boot-up.
Use this item to enable or disable OEM Logo. The default value is [Enabled].
When you select this option, the message “Save configuration changes and exit setup?” will pop-out. Press key or select [Yes] to save the changes and exit the UEFI SETUP UTILITY.
When you select this option, the message “Discard changes and exit setup?” will pop-out. Press key or select [Yes] to exit the UEFI SETUP UTILITY without saving any changes.
When you select this option, the message “Discard changes?” will pop-out. Press key or select [Yes] to discard all changes.
Load UEFI default values for all the setup questions. F9 key can be used for this operation.
PCH-FW Configuration
Storage Configuration
NVMe Configuration
ACPI Configuration
USB Configuration
Super IO Configuration
Serial Port Console Redirection
H/W Monitor
Runtime Error Logging
Intel SPS Information
Network Stack Configuration
VMD Configuration
Driver Health
Instant Flash
MCIO1-1/1-2/2-1/2-2 ASPM Support: Enables or disables the ASPM support for all CPU downstream devices.
ANSI
Extended ASCII character set
ANSI
Extended ASCII character set
VT100
ASCII character set
VT100+
Extended VT100 that supports color and function keys
VT-UTF8
UTF8 encoding is used to map Unicode chars onto 1 or more bytes
VT100
ASCII character set
VT100+
Extended VT100 that supports color and function keys
VT-UTF8
UTF8 encoding is used to map Unicode chars onto 1 or more bytes
12/31/2024
First release of Axial AX300 Series UEFI manual
AX300 Series Edge Server
The Axial AX300 Series of powerful, high-density edge servers are specifically engineered for complex computing workloads. Equipped with 4th 5th Gen Intel® Xeon® Scalable processing and a huge range of expansion and storage options, the AX300 Series delivers exceptional performance in a shallow depth, 3U form factor. Its industrial design and versatile installation capabilities make it the perfect fit for challenging edge computing environments. With incredible configuration flexibility, the AX300 Series is ideally suited for AI training and AI inferencing, virtualization, advanced automation, or any application that demands scalable, low latency computing capabilities at the edge.
For more information on accessories and additional features, visit the AX300 product page:
Install the device securely. Be careful handling the device to prevent injury and do not drop.
Equipment is intended for installation in a Restricted Access Area.
Ne pas ouvrir ou modifier l'appareil. L'appareil utilise des composants conformes aux réglementations FCC et EC. La modification de l'appareil peut annuler ces certifications.
Installez l'appareil en toute sécurité. Manipulez l'appareil avec précaution pour éviter de vous blesser et ne le laissez pas tomber.
The following accessories are included with every system:
Cable management ties (2RALXX220400)
3.5” bay mounting bracket (Pre-Installed with needed screws kit)
5.25” bay mounting brackets (2RALXX6751A1)
Spare M4 screws for slide rail (2RALXX6755A0)
Chassis keys
If additional items were purchased, such as rail mounting kits/brackets, they will be included in the system packaging.
The system label is located on the right side of the chassis as depicted in the image below. The system label will contain the following information:
System Model
OnLogic Serial Number
Regulatory & Compliance Certification Logos
On the front of the chassis, there is a retractable product information label containing pertinent product information such as:
System Model
OnLogic Serial Number
BMC MAC addresses
The ID LED/Button is available to assist with locating the system. ID may be turned On / Off by physically pressing the ID button. ID may also be turned On, Off, or set to Blink from the Baseboard Management Controller (BMC) Web UI.
There is one VGA Video port located on the back of the Axial AX300 Series Server. This is connected to the AST2600 BMC Chip.
There are 2x USB 3.1 Gen 1 Type-A ports in the front and 2x USB 2.0 Type-A in the rear of the Axial AX300 Series Server.
The AX300 Series features the following onboard Ethernet ports:
The 2 ports on the AX300 Series are 10GBASE-T ports
The AX300 Series features 7 PCI 5.0 Express slots on the motherboard (Note: PCIE1 is electrically x8). The PCIe lanes are routed to the listed CPU and offer support for various types of devices.
To maximize performance and compatibility of PCIe adapters, the following PCIe lane population order should be followed when adding PCIe adapters to the system.
PCIe lanes are numbered below for reference.
This diagram provides a comprehensive map of the AX300 motherboard, detailing various headers and illustrating the optimal installation of RAM and PCIe cards based on CPU and RAM module configurations
The M.2 Socket (M2_1, Key M) supports type 2280/22110 M.2 PCI Express modules up to Gen5 x4 (32GT/s x4).
There is one SATA header on the AX300 Series Server motherboard. The data ports support SATA III 6Gbps storage devices.
There are 2 MCIO headers on the Axial AX300 Series motherboard. Each supports up to PCIe5.0 x8. By default these connections are used for the 4x Front Drive Bay option and uses the CBDT119 cable (4x OCuLink to 2x MCIO)
There are three OCuLink headers on the Axial AX300 Series motherboard that each support PCIe3.0x4 or 4 SATA 6Gb/s per port.
When configured for SATA,
OCU 1 provides ports SATA0_0, SATA0_1, SATA0_2 and SATA0_3
OCU 2 provides ports SATA0_4, SATA0_5, SATA0_6 and SATA0_7
OCU 3 provides ports SATA1_4, SATA1_5, SATA1_6 and SATA1_7
Note that when configured as SATA, OCU 1 and 2 occupy a single VROC RAID instance apart from OCU 3, and therefore a RAID volume spanning all 3 OCuLink ports cannot be created.
OCuLink connector labeling:
All expansion storage options are located in either the 3.5” or 5.25” expansion bays:
5.25” Front Bay Options:
8x SATA SSD Bay - 8x 2.5", 5mm to 7mm, SATA Drives (8x SATA)
4x NVMe SSD Bay - 4x 2.5" U.2/U.3 (SFF-8643) NVMe SSD Drives (4x OCuLink)
12x M.2 Bay - 12x M.2 SATA Drives (3x OCuLink, 4 drives per connection)
3.5” Rear Bay Options:
1x SATA HDD/SSD Bay - 1x 2.5”, 5mm to 15mm, SATA HDD & SSD Drive (1x SATA)
The following drive population recommendations are provided to ensure consistency of connectivity, operation, and OS drive enumeration aligned to physical drive bay locations.
The front 8x SATA SSD Bay and rear 1x SATA HDD/SSD Bay option both feature locking mechanisms to prevent unauthorized drive removal.
The AX300 Series supports up to sixteen (eight per CPU) 288-pin DDR5 DIMM slots in two groups, and supports Single Channel Memory Technology.
5th Gen Intel Xeon Scalable Processors support transfer speeds between 4800MT/s and 5600MT/s depending on the processor SKU.
The IMC (Integrated Memory Controller) of the processors supports two mirror modes. The ability to support the memory modes is dependent upon the DIMM population. There are 2 IMC per CPU.
Mirror Modes
Full Mirror Mode will set the entire memory in the system to be mirrored, consequently reducing the memory capacity by half. One half will remain active while the other half will be left in reserve.
Partial Mirror Mode will enable the required size of memory to be mirrored. If rank sparring is enabled, partial mirroring of RAM will not take effect.
Enabling any type of Mirror Mode will disable XPT Prefetch.
ADDDC Sparing
ADDDC Sparing enables Memory Rank Sparing to reserve memory ranks to replace failed memory ranks when an error is detected. This will reduce the total memory available to the OS. This feature is disabled by default in the BIOS.
Patrol Scrub
Patrol Scrub is a background activity initiated by the processor to seek out and fix memory errors. This feature is disabled by default in the BIOS.
The following rules apply to when populating DIMMs in the AX300 Series Server:
Only DDR5 DIMMs may be installed into the system.
The maximum frequency of the system memory will never exceed that of the lowest frequency DIMM(s) installed in the system.
When populating DIMMs within the system, the following population is recommended for each CPU configuration in order to maximize overall system performance (V indicated populated slot):
Single CPU configurations:
Dual CPU configurations:
When GPUs are installed in the system, it is recommended that the minimum system memory should be at least 1.5 times the total GPU memory. For superior performance and to accommodate future workloads, it is strongly recommended that the system memory be 2.0 times the total GPU memory. Failure to meet these memory requirements may result in degraded performance or unexpected application behavior.
The Axial AX300 Series supports an optional discrete TPM 2.0 module. The TPM module included with the AX300 Series Server is the Infineon SLB9670 TPM2.0 (13-pin SPI module)
A socket is provided for a CR2032 battery. Regulatory requirements dictate the installed battery shall be rated for operation to at least 85°C.
A connector is provided for a cable connection to the power supply PDB.
A 10-pin (9 electrical) header is provided for cabled serial port, for diagnostics or legacy remote management.
Seven fan headers are included. Hotswap is supported at the motherboard level.
The front panel header provides power button and LED, reset button and hard disk activity connections. AX300 exposes the power button and LED on the chassis and leaves the others disconnected.
The AUX front panel header provides connections for some auxiliary features as indicated. The Locator LED and System Fault LED are exposed on the AX301 chassis front, and CASEOPEN detection is used to report the lid status.
System voltages are provided to the motherboard through a standard 24-pin ATX power connector.
Four (4) 12V Aux power inputs are present. All four inputs are used in units built by OnLogic and can only be used with the CBPW166 cable.
A VROC key header is included to enable Intel Virtual RAID on CPU and NCME/AHCI RAID on CPU PCIE.
An SMBus connection to the BMC is provided.
A USB 3.2 Gen 1 connection is cabled to the front of the chassis to enable the front USB ports.
Dr. Debug is used to provide code information, making debugging easier. Please see the charts below for Dr. Debug code information.
Click here for the directly from the motherboard supplier.
The sections below focus on the power features and capabilities of the AX300 Series Server.
The system supports two redundant power supplies, which may be 1000W/2400W depending on the selected mode and input voltage (110V/240V respectively). These power supplies are hot-swappable, meaning they can be replaced while the system is running without interrupting its operation.
The power input into both supplies must be the same as they will automatically adjust their power mode depending on the input voltage. 110V input will have a maximum output of 1000W and 240V input will have a maximum output of 2400W.
If you need to replace a failed power supply, simply remove the failed unit and insert a new one of the same wattage. The system will automatically recognize the replacement power supply and bring it online to restore redundancy.
IMPORTANT: When utilizing multiple 300W-350W or the maximum of seven 150W PCIe adapters (such as GPUs), the 2400W power supply mode is recommended on both power supplies due to momentary power spikes (exceeding the rated wattage of the adapters) that may occur. When these power spikes occur, the power consumption of the PCIe adapters combined with power draw of other system components may exceed the available power of the supplies.
The power supplies in this system are fully redundant in a primary/backup mode. This means that the two power supplies work in parallel, with one power supply acting as the primary source of power and the other as a backup.
In normal operation, the primary power supply is responsible for supplying power to the system, while the backup power supply remains idle. If the primary power supply fails, the backup power supply automatically takes over, ensuring that the system continues to receive power without interruption.
The power supplies are designed to work seamlessly together, with the primary power supply handling the majority of the load and the backup power supply providing additional power as needed. This redundancy ensures that the system can continue to operate even if one power supply fails, providing a high level of reliability for critical systems.
If a power supply failure occurs, the alerts will be presented via the Baseboard Management Controller (BMC), an audible alarm may occur, and the error LED will assert. If this happens, the failing supply can be serviced while the system remains operational on the backup power supply. Once the replacement power supply is installed, the system will automatically detect it and bring it online, restoring full redundancy.
This table represents the max power draw of a max configuration AX300 Series Server with two 6548Y+ CPUs, a selection of GPUs, and 1 of the 3 options for storage docks in the front.
Based on the total power consumption of the system, the following guidelines should be followed relative to the power supply input voltages, quantities, and line cords:
The Axial AX300 can be configured to turn on automatically when power is connected. This is useful for power outage recovery or if the unit is mounted in a hard to reach location. You can adjust Auto Power On settings by following the steps listed below.
Power on the system and press F2 a few times to access the BIOS
Navigate to Server Mgmt > BMC Tools
Locate Restore AC Power Loss setting
The Axial AX300 Series Edge Server is designed to operate and function across a wide temperature (5 to 40°C) and humidity range (8 to 85% RH non-condensing). The following sections describe the thermal and cooling capabilities of the system and its expected behavior in various conditions.
The Axial AX300 Series Edge Server has five 80 x 80 x 38mm system fans, which can be independently controlled and configured via the Baseboard Management Controller (BMC) relative to the supported system temperature sensors. The default fan duty and configuration settings have been validated to operate in accordance with the supported temperature range (up to 40°C). If the ambient operating temperature is tightly controlled, additional fan configuration optimizations may be manually adjusted to optimize acoustics and reduce power consumption. For additional information pertaining to manual fan configuration settings, please consult the .
The power supply fans operate independently and have their own closed-loop cooling algorithm.
By default, the fans are divided into 3 cooling zones: PCIe Expansion, CPU and Memory, and Power Supply and Storage. Internal baffling ensures CPU air flow is directed through the CPU heatsinks. A partial diverting duct directs some of the exhaust air from CPU0 around CPU1 and bypass air from the inlet to CPU1.
Note: In the case of a configuration with only 1 CPU installed, there will be no diverting duct installed in the chassis
In addition to the system fans, some PCIe and 5.25” bay devices will have their own independently controlled fans.
Temperature sensor data is available for many onboard temperature sensors.
The system is configured to operate in accordance with a closed loop thermal algorithm which accounts for components’ temperature maximums, reduced acoustics, lower power consumption, and optimal performance.
The fan zone assignments, default closed loop tables, and associated temperature sensors are outlined in this section.
Assigned Temperature Sensor: CPU1 Temp, CPU2 Temp
Assigned Fans: FAN5, FAN6
As per the default configuration settings, the system fans will increase duty cycle at 1% increments every second when the CPU temperature is at or above 80°C. When the temperature drops below 65°C, the system fan duty cycle will reduce by 1% every second.
Assigned Temperature Sensor: TEMP_GPUx, M.2 Temp
Assigned Fans: FAN3, FAN4
As per the default configuration settings, the system fans will increase duty cycle at 1% increments every second when the GPU temperature is at or above 70°C. When the temperature drops below 60°C, the system fan duty cycle will reduce by 1% every second.
Note: GPU temperature sensing is only supported with Nvidia professional grade or Intel Datacenter GPUs
Assigned Temperature Sensor: TRMB1 (Ambient Temperature Sensor)
Assigned Fans: FAN7
As per the default configuration settings, the system fan will increase duty cycle at 1% increments every second when the ambient temperature is at or above 60°C. When the temperature drops below 50°C, the system fan duty cycle will reduce by 1% every second.
The default system idle duty cycle is 15% and the default maximum duty is 95%.
Due to the chassis air inlet and outlet sizing, the fan performance does not marginally improve above 95% duty cycle, while fan noise increases.
Upon System Fan Failure or BMC Firmware Update, System Fans will ramp to maximum speed.
Acoustical performance of the default fan settings is provided in terms of three configurations: Entry, Typical, and Maximum. Configuration details are provided in Table 4.3.3.1 below. Each configuration has been tested according to ISO7779 & ISO9296 @ 23°C.
The default fan duty and configuration settings have been validated to operate in accordance with the supported temperature range (up to 40°C) as per the following test scenario and results.
The system performance was tested along a ramp with the following levels at two hours per level. During the test sequence numerous points throughout the system were monitored to ensure adequate cooling was provided to components in the system. The system was also tested 5ºC and 10ºC above its rated temperature range to help classify performance outside of the rated temperature range.
Summary: The Axial AX300 Series system configured with entry level hardware performed at full capacity under the workload across the full thermal range (0C-40°C) without the CPU or GPU throttling.
Summary: The Axial AX300 Series system built with a typical hardware configuration performs at full capacity under the workload up to 40°C. CPUs maintain the full workload across the full temperature sweep. The Nvidia RTX 4500 ADA remains below critical temperatures for the operational range and begins throttling above 40°C as critical temperatures are reached.
Summary: The Axial AX300 Series system built with a maximized hardware configuration performs at full capacity under the workload up to 40°C. CPUs maintain the full workload across the full temperature sweep. The four Nvidia RTX 6000 ADA cards remain below critical temperatures for the operational range and begin throttling above 40°C as critical temperatures are reached.
CPUs with high Thermal Design Power (TDP) ratings may experience performance throttling under sustained, maximum workloads, particularly in high ambient temperature environments.
To optimize performance in high ambient temperature environments, it is recommended to characterize workloads accordingly and adjust system settings as needed.
The Axial AX300 Series Edge Server has been designed with flexibility in mind and can be mounted in different ways. As the system is designed to fit industry standard 19” Electronic Industries Alliance (EIA) racks, there are multiple rack mounting rail kits available. Additionally, a 19” EIA two-post rack mounting option is available.
Two options are available for mounting outside of a traditional rack. The system may be wall mounted using the OnLogic wall mount kit. Alternatively, the system supports feet for a vertical desktop configuration.
In addition to the mounting options, the Axial AX300 Series Edge Server supports locking front and rear security bezels.
The Axial AX300 Series Edge Server has been designed to support standard 19" EIA rack mounting, which is a common form factor used in data centers and server rooms. To accommodate different rack depths, 4 different rail options support 16.5" - 47" rack depths. These rail kits can be used to securely mount the server in the rack, are easy to install, and include all the necessary hardware for attachment into the rack.
All rail kit options are also supported in a “reverse” orientation, with I/O and power supplies facing the front of the rack.
The Ball Bearing Slide Rails are an optional accessory designed to enhance the functionality and ease of use of the Edge Server. These slide rails are designed to be used with standard 19" EIA racks and allow for easy installation and removal of the server from the rack. The ball bearing design ensures smooth and effortless motion, while the sturdy construction provides a secure and stable platform for the server. With these slide rails, you can easily access the server for maintenance or upgrades without the need for complex disassembly or cumbersome lifting.
The desired length Ball Bearing Slide rail kit can be chosen at time of configuration based on the rack depth requirements.
Mounting Hole: Square, Rack Depth Range (front to back flange): 420mm (16.5in) to 940mm (37in)
Forward Orientation: Install the six M4x0.7 L=4mm Low Profile Cheesehead screws provided with the rail kit. Align the first hole in the rail with the hole label specified in the table below.
Reverse Orientation: Install the six M4x0.7 L=4mm Low Profile Cheesehead screws provided with the rail kit. Align the first hole in the rail with the hole label specified in the table below.
The Cable Management Arm Slide Rail Kit (Part Number: 3RAMIS205300) is an optional accessory that enhances the standard ball bearing slide rails by providing a cable management arm to neatly organize and secure cable connections to the Edge Server system while still supporting easy removal of the server from the rack for maintenance and upgrades.
The Cable Management Arm is not compatible with the 420mm (16.5”) to 600mm (23.5”) slide rail.
For slide rail options 23.5” and up, the Cable Management Arm can be chosen at time of configuration based on the rack depth requirements.
Mounting: Integrated with slide rail
Clip the cable management arm into the brackets on the slide rail
The Two-Post Rack Bracket Kit (Part Number: MTR-2POST-AX301) is an optional accessory designed to enable flexible mounting of the Edge Server. These brackets are designed to be used with standard 19" EIA Two-Post racks and allow for stable and secure mounting of the server in a rack.
The Two-Post Rack Bracket Kit supports both forward and reverse mounting. To convert to a reverse mounting orientation, remove the handle.
The Kit includes all required hardware for square or M5 threaded racks. Additional hardware may be required for racks with other mounting holes.
Mounting Hole: Square or M5 threaded with included hardware. #10-32, #12-24 or round hole with appropriate additional hardware.
Rack Post Range (front to back flange): 50mm (2in) to 150mm (6 in)
Step 1: Install twelve M5 cage nuts in the rack as shown. Secure the four brackets to the rack posts using twelve M5x0.8 L=15mm Pan Head Screws. Install two M5 cage nuts in the front brackets.
Step 2: Install the edge server in the brackets. Secure the server using two M5x0.8 L=15mm Pan Head Screws.
The Axial AX300 Series Edge Server is designed to accommodate a reversed, front I/O rackmount orientation, with the I/O and power supplies facing the front of the rack.
All rail kit options with supported rear I/O are also supported in a “reverse” orientation, with I/O and power supplies facing the front of the rack. To convert to a reverse orientation, remove the rack handles and reattach them at the rear of the chassis.
Reverse Orientation: Install rack handles at the rear of the chassis using eight #6-32 L=1/4” Pan Head Screws.
Airflow Implications: Most racks have a front-to-back airflow expectation, where cooling air is supplied at the front and exhaust air is removed at the rear of the rack. Mounting a standard AX300 Series server in the reverse configuration, without accommodations for reversing airflow, will violate this convention and may cause cooling problems for either the AX300 or other devices in the rack.
Options to enable front-to-back airflow with front I/O orientations are available. Consult OnLogic for additional information.
The Axial AX300 Series Edge Server wall mount kit (Part Number: MTW116) is made of sturdy metal and designed to securely hold the server in place against a wall. This optional accessory includes the necessary wall mounting brackets and hardware to flexibly mount the Axial AX300 Series Edge Server system where a rack is not available or practical. It is strongly recommended to include the front security bezel and rear cable security bezel when wall mounting the edge server. If installed, the front security bezel can also support an optional dust filter.
Install the eight M4x0.7 L=4mm Flathead screws provided with the wall mount kit. Use holes “F1” and “B1” for the outermost holes on the bracket. Install four plugs in holes marked with “T” or “∆”. (right side only)
The optional tower feet (Part Number: MTT-AX301) allow the Axial AX300 Series Edge Server to be used vertically in a desktop tower configuration. Using the front and rear security bezels in conjunction with the tower feet is strongly recommended.
Install four tower feet using one M4x0.7 L=6mm Pan Head screw each, use holes marked “T” or “∆” on the right side of the chassis. Install nine plugs in slide rail holes “F2” through “B2” (right side only)
Stand the system onto the four tower feet.
The Axial AX300 Series Edge Server offers an optional front security bezel (Part Number: F1-AX301) that prevents access to the front USB ports, buttons and hot-swap chassis fans. For rack mounted configurations, the front bezel also prevents access to the rack screws to prevent unauthorized removal from the rack. In addition to the security features, the front bezel also contains a replaceable dust filter.
Replacement dust filters are available:
The rack-out protection prohibits the removal of the system through the tabs on the bottom corners, offering an additional layer of security.
If the rack-out protection is not desired, the screw covers can be removed. Press firmly upward on the screw cover until the pegs pop out of the keyhole slots.
Install the bezel bracket with three M3 flat head screws.
Angle the bezel and align the hooks on the right side of the bezel with the bezel bracket. Using one hand to keep the right side of the bezel in place, press the bottom left corner inward and upward until the hook pops into place. Use the key to lock the bezel.
The rear cable security bezel (Part Number: B2-AX301) for the Axial AX300 Series Edge Server prevents unauthorized removal or installation of cables or devices from the rear I/O ports. The bezel has openings on the left and right sides to allow cables to pass through. These openings are protected with dust brushes, and the rear bezel supports an optional dust filter for use in reverse airflow configurations.
Use the included five #6-32 Flat Head T10 Screws to attach the base of the rear security bezel.
The Axial AX300 Series is certified to NEBS GR-63-CORE Earthquake Zone 4 Shelf-Level component. To achieve this rating, the use of locking cables for all connections is required. Many I/O connections have a locking mechanism as part of the connector specification. For these connections, use of a specification compliant cable is sufficient to meet the requirement.
Locking power cables can be purchased from Onlogic using the SKUs below:
The Axial AX300 Series Edge Server is designed to be compact while maintaining easy serviceability. Please follow the instructions below to service the specified component or device.
Important Note: Except for hot-swapping the chassis fans/power supplies, the system should be powered off and the power disconnected before performing any service.
To open the AX300 you will want to Loosen the single retention screw on the system’s cover and unlock the Chassis Lock. You can then pull the chassis lid back and access the units internal components.
The 5 chassis fans on the front of the system are hot swappable.
Step 1: Grip the handle and the release latch on the front of the fan cage. Step 2: Depress the release latch and pull straight out to remove the fan. Step 3: Insert the replacement fan into the socket. Press the fan until the latch clicks. Check that the replacement fan is secure by gently pulling the handle.
Step 1: Unlock and remove the Front Security Bezel. Step 2: Remove the dust filter from the bezel. The filter frame is flexible and can be removed by gently lifting one corner until the retaining tabs disengage. Step 3: To install a new filter, simply press the filter into the bezel. Ensure all 8 retaining tabs are engaed:
Step 1: Unlock and open the chassis lid. Step 2: Disconnect data and power cables on the cable routing bracket at the PSU side. The other end of these cables can be left connected to the motherboard or PCIe devices as appropriate Step 3: Remove (or fold to one side) the cable routing bracket. Step 4: Remove the 3.5” bay bracket and 5.25” bay device (if populated).
Step 5: Remove the #6-32 Flange Head Screw securing the CPU air duct. Step 6: Bend the CPU air duct material to unlatch the opposing hooks at the back left corner. Step 7: Remove the right side (L-shaped piece) of the CPU air duct.
Step 8: Remove air diverter from between the CPU sockets. Step 9: Service CPU or Memory devices. Tip: Access to memory release levers can be improved by removing chassis FAN 5 or FAN 6.
To remove the CPU:
Step 1: Using a T30 bit, loosen the 4 corner fasteners on the cooler in a star pattern to ensure even pressure release. Step 2: Pull the wire clips up toward the CPU to unlatch the CPU cooler. Step 3: Gently pull the CPU cooler up by the edges as the CPU is connected to the cooler. Step 4: Once cooler is removed, the CPU carrier needs to be removed from the cooler by unlatching it gently. Step 5: Remove the CPU from the CPU carrier by lifting the latch release handle found on the bottom of the carrier.
To replace the CPU:
Step 1: Line the CPU up on the CPU carrier using the golden triangle found on the corner of the CPU and additionally using the notches found on the CPU and CPU carrier. Step 2: Push down metal handle to lock CPU into the carrier. Step 3: Flip CPU carrier over and apply thermal paste to the top of the CPU. Step 4: Clip the CPU carrier onto the CPU cooler with the TOP of the CPU making contact with the BOTTOM of the cooler. Keep in mind the corner of the CPU that contains the triangle for alignment.
Step 5: Line up the cooler + CPU carrier with the socket on the motherboard making sure the screws align with the stand offs found on the motherboard and the triangle lines up with the triangle found in the corner of the socket.
Step 6: Flip clips found on the cooler fasteners down, away from the CPU to latch the cooler to the board. Step 7: Using a T30 bit, tighten the CPU cooler fasteners to the motherboard in a star pattern to ensure even distribution of pressure
Step 1: Replace the left wall of the air duct immediately to the left of the outermost RAM slot as shown below. Step 2: Replace air diverter between the CPU sockets. Connect the hook into the corresponding notch in the left wall of the air duct. Note orientation label on the air diverter. Step 3: Replace air duct right wall in chassis. Each wall of the air duct should be inserted just outside the outermost RAM slot, such that all RAM slots are inside the duct. See below:
Step 4: Align hooks on left side of air duct, then bend duct material to latch rear most opposing hook.
Step 5: Install the #6-32 Flange Head Screw to secure the CPU air duct. Step 6: Reinstall cable routing bracket and connect cables. Step 7: Reinstall 3.5” and 5.25” devices. Step 8: Close and lock lid.
To remove DIMMS:
Step 1: Press down on the latches on the DIMM slots on the motherboard until they are tilted away from the DIMM. Step 2: Gently pull the DIMM out of the slot.
To replace DIMMS:
Step 1: Press down on the latches on each side of the DIMM slot on the motherboard until they are tilted away from the DIMM slot. Step 2: Line up the notch in the DIMM with the notch in the slot to ensure the DIMM goes in correctly. Step 3: Gently press the DIMM into the slot until an audible click is heard which indicates the latches have been engaged and are holding the DIMM in place.
Step 1: Replace the left wall of the air duct immediately to the left of the outermost RAM slot as shown below. Step 2: Replace air diverter between the CPU sockets. Connect the hook into the corresponding notch in the left wall of the air duct. Note orientation label on the air diverter. Step 3: Replace air duct right wall in chassis. Each wall of the air duct should be inserted just outside the outermost RAM slot, such that all RAM slots are inside the duct. See below:
Step 4: Align hooks on left side of air duct, then bend duct material to latch rear most opposing hook.
Step 5: Install the #6-32 Flange Head Screw to secure the CPU air duct. Step 6: Reinstall cable routing bracket and connect cables. Step 7: Reinstall 3.5” and 5.25” devices. Step 8: Close and lock lid.
To remove DIMMS:
Step 1: Press down on the latches on the DIMM slots on the motherboard until they are tilted away from the DIMM. Step 2: Gently pull the DIMM out of the slot.
Step 1: Unlock and open the chassis lid. Step 2: If a 3.5” Device is already installed, disconnect power and data cables. Step 3: Remove the 3.5” bay bracket from the chassis. Loosen the thumbscrew, then pull the bracket backward into the chassis before lifting upward. Step 4: If no device is installed in the bay, turn the bracket upside down and remove the two screws holding the blank plate onto the bracket. (Figure 2 below)
Step 5: Insert the new 3.5” device at a slight angle onto the 2 pins on the right side of the bracket. Secure the device using M3 screws. (Figure 3 above) Step 8: To install the bracket into the chassis, align the slot pins on the underside of the bracket with the corresponding holes in the chassis. Then slide the bracket forward and secure the thumbscrew. The front of the device should be flush with the rear of the chassis. Step 9: Connect power and data cables to the device.
Step 1: Unlock and open the chassis lid. Step 2: If a 5.25” Device is already installed, disconnect power and data cables, then proceed with Step 3a. If the 5.25” blank cover is installed, skip to step 3b. Step 3a: Remove the 5.25” bay bracket from the chassis. Loosen the thumbscrew and pull the tab on the release pin, then pull the bracket backward into the chassis before lifting upward. Step 3b: If no device is installed in the bay, remove the empty bracket by pulling the two levers on the release pins:
Step 4: Locate the 5.25” brackets in the accessory box, then secure them to the sides of the device using M3 screws as shown below:
Step 5: To install the bracket into the chassis, align the slot pin on the underside of the bracket with the corresponding hole in the chassis. Then slide the bracket forward and secure the latch pin and thumbscrew. The front of the device should be flush with the front of the chassis. Step 6: Connect power and data cables to the device. Set the fan speed switch to desired “LOW” or “HIGH” speed indicated by “L” and “H” located at the bottom right of the bay.
To help protect large PCIe devices (such as GPUs) in high vibration environments, the server supports the addition of bracket(s) to support the back end of the device. Depending on configuration, the system may have a bracket for each card, or a single bracket supporting all installed full-height/full-length devices.
Note: Half-length/low profile PCIe devices do not have a bracket
Step 1: Unlock and open the chassis lid. Step 2: If installing a new device, start by removing the PCIe blank plate. Step 3: Attach the bracket to the GPU using M3 screws. Bracket dimensions and screw location will vary depending on the GPU. An example installation is shown below. Consult Onlogic for supported GPU brackets.
Step 4: Install the GPU and bracket assembly into the chassis. Secure with #6-32 flange head screws.
Step 1: Unlock and open the chassis lid. Step 2: If installing a new device, start by removing the PCIe blank plate. Step 3: If the PCIe Support Bracket is already installed, remove it by removing the four #6-23 flange head screws. Step 4: Install GPU into motherboard, secure front of card with #6-32 flange head screw(s). Step 5: Prepare the bracket: for dual slot cards, twist out the appropriate dividers using a screwdriver. Note: Chassis/bracket PCIe slot numbering/order may not align with the motherboard. Step 6: Install the bracket into the chassis. If the dividers get stuck gently rock the GPUs until the divider fits into the gap between the cards. Secure with four #6-32 flange head screws.
For complete details on BIOS/UEFI configuration, refer to the official BIOS/UEFI User Manual:
For complete details on the Baseboard Management Controller (BMC) functionality, refer to the official BMC Manual:
Click here for the directly from the motherboard supplier.
Update the BIOS with the downloaded file(s) above. Refer to the AX300 BIOS Manual (linked above), Please use Instant Flash for update procedure.
The AX300 Series Server supports the following operating systems:
Microsoft Windows Server 2022 Essentials
Microsoft Windows Server 2022 Standard
Ubuntu 22.04 - Server
Ubuntu 24.04 - Server
RAID (Redundant Array of Independent Disks) is a technology that allows multiple hard drives to work together as a single logical drive, providing increased performance and data redundancy. The idea behind RAID is to combine the storage capacity of multiple drives to create a larger virtual drive that appears to the operating system as a single disk.
RAID can improve system performance by distributing data across multiple drives, allowing for faster read and write speeds. Additionally, RAID can provide data redundancy by using multiple drives to store the same data, so that if one drive fails, data can still be accessed from the other drives. There are several RAID levels with different configurations and benefits, each offering varying levels of performance and data redundancy.
The Axial AX300 Series Server supports onboard RAID via Intel Virtual RAID on CPU (Intel VROC) supported by the Intel Xeon Scalable processors.
Intel VROC is an enterprise RAID solution that unleashes the performance of NVMe SSDs, enabled by a feature in Intel Xeon Scalable processors called Intel Volume Management Device (Intel VMD), an integrated controller inside the CPU PCIe root complex.
Prior to configuration of RAID, users are advised to back up their data as the process may erase all data on the hard drives.
Note: SATA RAID is limited to volumes from disks on the same controller. SATA0_0 through SATA0_8 (OCuLink ports 1 and 2) may be used to create up to an 8-disk RAID volume, and SATA1_4 through SATA1_7 (OCuLink port 3) may be used to create a separate, up to 4-disk RAID volume. This limitation does not apply to NVMe RAIDs created using the OCuLink ports.
The following sections will discuss the various RAID types that are supported on the AX300 Series Server and their respective advantages/disadvantages.
Intel VROC allows for RAID volumes to be created and controlled by the Intel Volume Management Device (Intel VMD) controller. There are 2 VROC options that can be purchased with the system:
VROC Standard - Allows for RAID 0/1/10 volumes to be created
VROC Premium - Allows for RAID 0/1/5/10 volumes to be created
RAID 0 (Redundant Array of Inexpensive Disks level 0), also known as striping, is a method of combining multiple physical hard drives into a single logical volume for improved performance.
In RAID 0, data is divided into blocks and spread across two or more physical drives simultaneously. The blocks are written to the drives in a way that balances the load and optimizes performance. When data is read, the blocks are retrieved from multiple drives at the same time, increasing the read and write speed of the overall system.
An advantage of RAID 0 is its improved performance due to the parallel access to multiple drives. However, RAID 0 does not provide any fault tolerance or redundancy. If one drive fails, the entire RAID 0 volume will be lost, along with all data stored on it. Therefore, it is recommended to use RAID 0 only for non-critical data or as part of a larger backup and disaster recovery strategy.
RAID 0 requires a minimum of two drives..
For RAID 0, it is recommended to use disks of the same interface, speed, and capacity. If the disks in a RAID 0 array have different sizes, performance may be limited and the capacity of the array will be limited by the size of the smallest disk.
RAID 1 (Redundant Array of Inexpensive Disks level 1) is a type of data storage technology that provides data redundancy and fault tolerance by creating an exact copy, or mirror, of data on two or more physical drives.
In RAID 1, when data is written to one drive, it is simultaneously written to the other drive(s), creating an exact duplicate of the data on each drive. This ensures that if one drive fails, the data can still be accessed from the remaining drive(s). The read performance of RAID 1 can be faster than that of a single drive because data can be read from multiple drives at the same time. However, write performance is typically slower because data must be written to multiple drives.
An advantage of RAID 1 is its data redundancy and fault tolerance. If one drive fails, the data is still available on the other drive(s). Additionally, RAID 1 can be hot-swappable, meaning that if a drive fails, it can be replaced without having to shut down the system.
However, RAID 1 has some disadvantages, including lower storage capacity compared to other RAID configurations and higher cost due to the need for multiple drives. RAID 1 is recommended for applications that require high data availability and reliability, such as mission-critical systems, servers, and database applications.
RAID 1 requires a minimum of two drives.
For RAID 1, it is recommended to use disks of the same interface, speed, and capacity. If the disks in a RAID 1 array have different sizes, performance may be limited and the capacity of the array will be limited by the size of the smallest disk.
RAID 5 (Redundant Array of Inexpensive Disks level 5) is a type of data storage technology that uses striping with distributed parity.
In a RAID 5 configuration, data is striped across multiple disks, with parity information distributed across all the disks. This provides fault tolerance and redundancy, allowing data to be reconstructed in the event of a single drive failure.
RAID 5 offers good performance and fault tolerance for small to medium-sized businesses, but it has a higher overhead and is more complex than some other RAID configurations. Additionally, in the event of a second drive failure, data loss can occur. RAID 5 is often used in applications that require a balance between performance, fault tolerance, and cost.
RAID 5 requires a minimum of three disks, and the capacity of one disk is used for parity information.
For RAID 5, it is recommended to use disks of the same interface, speed, and capacity. If the disks in a RAID 5 array have different sizes, performance may be limited and the capacity of the array will be limited by the size of the smallest disk.
RAID 10 (Redundant Array of Inexpensive Disks level 10), also known as RAID 1+0 or mirrored striped volumes, is a combination of RAID 1 and RAID 0. It provides both data redundancy and improved performance.
In a RAID 10 configuration, multiple pairs of disks are configured as RAID 1 arrays, where data is mirrored between each pair of disks for redundancy. The resulting RAID 1 arrays are then striped together in a RAID 0 array, where data is striped across all of the mirrored pairs for increased performance.
Data is striped across the mirrored pairs, so the capacity of the RAID 10 array is equal to half of the total capacity of the disks. For example, in a four-disk RAID 10 array with 1TB disks, the total capacity of the array would be 2TB.
RAID 10 provides both performance and redundancy benefits, as it offers the performance benefits of RAID 0 while also providing the redundancy of RAID 1. In the event of a single disk failure, the mirrored pair can continue to provide access to the data. However, if both disks in a mirrored pair fail, data may be lost.
RAID 10 requires a minimum of four disks, and must have an even number of disks.
For RAID 10, it is recommended to use disks of the same interface, speed, and capacity. If the disks in a RAID 10 array have different sizes, performance may be limited and the capacity of the array will be limited by the size of the smallest disk.
RAID volumes can be configured and created via the BIOS or from an operating system (OS).
If an operating system is to be installed on to a RAID volume, the processes outlined in this section must be followed in order to appropriately enable RAID and create the RAID volume where the OS will be deployed.
This section will outline the process for creating RAID volumes outside of the OS via the BIOS.
Prior to configuring or creating any RAID volumes using Intel Virtual RAID on CPU, Intel Volume Management Device (VMD) must be appropriately configured/enabled.
From UEFI System Setup, navigate to Advanced → Intel VMD Technology → VMD Config for PCH ports → set to Enabled. New options will appear
Next, configure the VMD Enabled devices to be enabled:
Enable/Disable VMD → Enabled
After VMD has been enabled, a RAID volume can be created. The following procedure outlines the process to create a RAID volume using the RAID configuration utility:
Press F2 or Del to enter UEFI System Setup and navigate to Advanced → Intel(R) VROC SATA Controller
Select "Create RAID Volume" to create a new RAID volume.
Assign a Name and Select the RAID Level you want to create (e.g. RAID 0, RAID 1, RAID 5, etc.) and specify the settings for the RAID volume (e.g. strip size, capacity, etc.).
Choose the hard drives you want to include in the RAID array and add them to the volume by assigning them with an X.
Select Create Volume
It's important to note that deleting a RAID volume will erase all data on the hard drives in the array, so be sure to back up any important data before proceeding. The specific steps to delete a RAID volume may vary depending on the RAID configuration utility used and the RAID level in use.
To delete a RAID volume, follow these steps:
During the system boot-up process, press "Ctrl+I" to enter the RAID configuration utility.
Select the RAID volume you want to delete and choose the "Delete RAID Volume" option.
Confirm that you want to delete the RAID volume.
Save the changes and exit the RAID configuration utility.
RAID volumes can be created, configured and managed from within Windows. This section will outline the requirements and processes for doing so.
Note: Enabling VMD Configuration and Creating a RAID Volume in BIOS steps are prerequisites for installing an OS to a RAID volume.
To install an OS on to a created RAID volume, perform the following steps to install the Intel Virtual RAID on CPU driver during operating system setup:
Download the latest Intel Virtual RAID on CPU Driver package from the OnLogic Support Site or and extract the contents to a USB drive.
Connect the USB drive to the computer where you want to install Windows.
Power off the system.
These steps and more are provided by Intel here:
Prior to configuring a RAID volume within the Windows OS environment, it is necessary to download the required drivers. The following procedure will outline the steps to ensure the proper drivers are downloaded and installed:
Download the Intel Virtual RAID on CPU software from OnLogic or Intel’s website.
Save the file to a known location on your computer's hard drive.
The following document outlines the procedure for creating a new RAID volume within the Intel Virtual RAID on CPU application from the operating system:
Open the Intel Virtual RAID on CPU application.
Click the “Create” icon to create a RAID array.
In “Select Volume Type”, click the desired RAID configuration. Click “Next”.
In “Configure Volume”, select the RAID disks then click “Next”.
After creation of the volume, to make the RAID volume usable from within the OS, it will need to be initialized, partitioned, and formatted (similar to a standard physical disk). To do so, follow the procedure below:
From the Windows Disk Management application, initialize the disk (the newly created RAID volume) such that Logical Disk Management can access it.
Right-click on the Disk associated with the RAID Volume and select “New Simple Volume”
Follow the instructions on the New Simple Volume Wizard.
After the volume wizard process is completed, the RAID volume should now be operational and the RAID volume will appear as if it were a single storage drive.
The following process outlines the procedure for deleting a RAID volume within the Intel Virtual RAID application from the operating system.
Open the Intel Virtual RAID on CPU application.
Click the “Manage” icon.
Select the RAID volume that is to be deleted.
Select “Delete Volume”
Warning: Deleting a RAID volume will destroy all contents held within the RAID array.
Intel VROC for Linux is mostly delivered through open-source operating system kernel and user space tools, with no additional software download required for specific Linux* distribution releases. It is up to the specific operating system vendor to pull-in Intel VROC features and patches. The distributions below have Intel VROC support, with newer releases being more complete.
Intel Virtual Raid on CPU (Intel VROC) in Linux Support Page
Additionally, as the configuration and implementation details for Intel VROC RAID in Linux may vary between distributions, please refer to the additional documentation below:
Red Hat Enterprise Linux 8 - Managing RAID
Red Hat Enterprise Linux 9 - Managing RAID
OS Support List:
What is BMC, and what is it for?
General information about the BMC, or Baseboard Management Controller, are discussed on .
For the latest security advisories concerning OnLogic products, including vulnerability disclosures and necessary updates, please refer to our official Security Advisories page. It is recommended to regularly check this resource for critical security information.
The Axial AX300 Series Edge Server supports an optional front security bezel. The security bezel helps prevent unauthorized access and tampering with the front ports and buttons of the system.
A barrel lock is used to secure the security bezel in place. A key for the barrel lock is included in the accessory package. The key is shared with the top lid lock and the rear cable bezel lock.
In the event that the bezel is removed while power is present to the system, the bezel intrusion switch will detect this event and the Front Bezel Intrusion sensor will be asserted. Relative to the intrusion, this event will also be logged in the Baseboard Management Controller event log.
The Axial AX300 Series Edge Server chassis lid has a two point locking mechanism with an intrusion mechanism built natively into the system chassis.
For the two point locking mechanisms, the first locking point is the top barrel lock. A key for the top barrel lock is included in the accessory box, and is shared with the front security bezel and the cable bezel.
The second lid locking point is a thumb screw located in the rear of the system.
In the event that the system lid is removed while power is present to the system, the intrusion switch will detect this event and the Chassis Intrusion sensor will be asserted. Relative to the intrusion, this event will also be logged in the Baseboard Management Controller event log.
The Axial AX300 Series Edge Server supports an optional rear cable bezel (Part Number: B2-AX301). The cable bezel helps prevent unauthorized access and tampering with the rear ports, power supplies, buttons and cables of the system. The lid has brush pass throughs on either side to allow routing of connected cables.
The rear cable bezel is composed of two pieces: a base that is screwed to the main chassis and a lid that is secured with a 3 point latch. A key for the latch is included in the accessory package. The key is shared with the top lid lock and the front security bezel lock.
Do not open or modify the device. The device uses components that comply with FCC and CE regulations. Modification of the device may void these certifications.
The use of shielded cables for connection of a monitor to the GPU is required to assure compliance with FCC and CE regulations.
The computer system was evaluated for IT equipment EMC standards as a class A device. The computer complies with the relevant IT equipment directives for the CE mark. Modification of the system may void the certifications. Testing includes: EN 55032, EN 55035, EN 60601-1, EN 62368-1, EN 60950-1.
This device complies with part 15 of the FCC rules as a Class A device. Operation is subject to the following two conditions: (1) this device may not cause harmful interference and (2) this device must accept any interference received, including interference that may cause undesired operation.
This device complies with Industry Canada license-exempt RSS standard(s). Operation is subject to the following two conditions: (1) this device may not cause interference, and (2) this device must accept any interference, including interference that may cause undesired operation of the device.
Le présent appareil est conforme aux CNR d'Industrie Canada applicables aux appareils radio exempts de licence. L'exploitation est autorisée aux deux conditions suivantes: (1) l'appareil ne doit pas produire de brouillage, et (2) l'utilisateur de l'appareil doit accepter tout brouillage radioélectrique subi, même si le brouillage est susceptible d'en compromettre le fonctionnement.
CAN ICES-003(A) / NMB-003(A)
The computer system was evaluated for medical, IT equipment, automotive, maritime and railway EMC standards as a class A device. The computer complies with the relevant IT equipment directives for the UKCA mark
This is a Class A product based on the standard of the Voluntary Control Council for Interference (VCCI). If this equipment is used in a domestic environment, radio interference may occur, in which case the user may be required to take corrective actions.
Chipset
Intel C741
Integrated Graphics
VGA Graphics through AST2600 BMC
Front I/O
2x USB 3.2 Gen 1 Type-A
1x Power Button / LED (White)
1x ID button / LED (Blue)
Rear I/O
1x Error LED (Orange)
1x 1GbE Dedicated Management (BMC/IPMI)
2x 10GbE LAN, Intel X710-AT2
2x USB 2.0 Type-A
1x VGA (BMC Video Output)
1x ID button / LED (Blue)
1x COM DB9
Expansion & Storage
1x M.2 2280/221100 M-key (PCIe Gen 5 x4)
6x PCIe Gen 5 x16 / CXL 1.1 Full Height, Full Length Slot
1x PCIe Gen 5 x8 / CXL 1.1 (x8 Electrical, x16 Compatible) Slot
2x MCIO (PCIe5.0 x8) [CPU0]
3x OCuLink (PCIe3.0 x4 or 4 SATA 6Gb/s) [PCH]
Up to 13 SATA 6Gb/s, support RAID 0/1/5/10
All AX300 Series Servers
1x 2.5" SATA Rear Drive Bay
AX301, AX303, AX305, AX307
Special Features
ASPEED AST2600: Full Web UI, iKVM, vMedia support
Optional TPM 2.0 module (Infineon SLB9670) or Intel PTT (Native)
Chassis Intrusion Detection
Security Bezel
Secure Boot
Hot Swap Fans
Bezel Detection
Operating Systems
Microsoft Windows Server 2022 Essentials
Microsoft Windows Server 2022 Standard
Ubuntu 22.04 - Server
Ubuntu 24.04 - Server
Red Hat Enterprise Linux 8.10 (or newer)
Red Hat Enterprise Linux 9.x
VMware vSphere and ESXi 8
LAN Controllers
1x Intel X710-AT2 Controller (2 x 10GBASE-T ports on Rear)
1x GbE Dedicated Management (BMC/IPMI)
Power Supplies
Up to 2 PSUs with PMBUS monitoring, 120~220 VAC 50-60Hz input
Delta 1000W/2400W CRPS 80 Plus Titanium AC Power Supply Forward Airflow (110V/220V Autoranging)
Dimensions (W x H x D)
438 x 132 x 421.2mm (17.3 x 5.2 x 16.6”) without Security Bezel
482.6 x 132 x 455mm (19.0 x 5.2 x 17.9”) with Security Bezel & Handles
132 x 448.2 x 553.6mm (5.2 x 17.6 x 21.8”) with Tower Stand, Security Bezel and Rear Cable Security Bezel
Weight
System Maximum: 24 kg (52.9 lbs)
Shipping Maximum: 35 kg (77.2 lbs)
Mounting
2-post 19” EIA Mounting Bracket
Wall Mount
Tower Stand
Rack Mount 17” (430-600mm) slide rail
Rack Mount 23” (564-850mm) slide rail with Optional Cable Management Arm
Rack Mount 28” (626-940mm) slide rail with Optional Cable Management Arm
Operating Temperature
5-40°C (ASHRAE A3 Operating Temperature) (Configuration Dependent)
Maximum ambient temperature decreases by 1°C for every 175m (574 ft) increase in altitude above 900m (2,953 ft)
Storage Temperature
-40-70°C
Operating Humidity
8~85% Relative, non-condensing
Maximum dew point 24°C
Storage Humidity
0~95% Relative, non-condensing
Maximum dew point 24°C
Acoustic
~50 dBA Typical Operation (See Section 4.3.3 "Fan Default Acoustics")
Earthquake (Shock & Vibration)
NEBS GR-63-CORE Zone 4 Resistance [Shelf-Level GR-63 4.4.1]
Package Test
ISTA 6-FEDEX-A
IEC/EN/UL 62368-1 (UL File No. E490677)
Others
Other countries may be available, contact us to learn more.
Power Off: The system will remain off when power is restored
Last State: The system will recover to the state it was in before the power loss event (i.e. If the unit was off, it would stay off. if the unit was powered on, it would power back on.)
Power On: The system will power on after any power loss event
Press F10 to Save & Exit
79C
92C
56C
Max Configuration
87C
94C
60C
B2
No
3RAMIS205200
564mm (22.2”)
850mm
(33.4”)
F1
B1
Yes
3RAMISBBG003
626.4mm (24.7”)
939.9mm (37.0”)
F1
B1
Yes
Red Hat Enterprise Linux 8.10
Red Hat Enterprise Linux 9.x
VMware vSphere and ESXi 8
Enable Hot Pluggable if desired
Go into Intel VMD for Volume Management Device on Socket 0/1
Enable VMD on PCIE 0/1/2/3/4/5/6 and MCIO1/2 as needed
Press F10 to Save and Exit. The system will then reboot.
Reboot the system and verify that the RAID array has been detected by the operating system or OS installation media.
Reboot the system and verify that the RAID volume has been deleted.
When the system starts, press F11 to bring up the boot menu and select the option to boot from the Windows installation media.
When the Windows Setup screen appears, press the "F6" key to install third-party RAID drivers or use the “Load Driver” option to load the F6 drivers.
Windows Setup will prompt you to insert the driver disk for the RAID controller. Insert the USB drive containing the RAID driver package and click "OK".
Windows Setup will scan the USB drive and display a list of compatible RAID drivers. Select the appropriate driver for the RAID controller (e.g. Intel RAID on Chip) and click "Next".
Windows Setup should now detect the created RAID volume(s) and allow for installation of Windows onto them as if they were a singular physical disk.
Continue with the Windows installation as usual.
Click Continue (if needed) to launch the installation program.
Click Next at the Welcome screen.
After reading and reviewing the warnings, Click Next.
Read the license agreement. To agree and proceed, click Yes to accept the terms and continue.
From the Readme file information, Click Next. The application files will now be installed.
When the appropriate installation files have been installed, you will be prompted to Click Next to continue.
Click Yes to the restart option and then click Finish to restart the system.
After restarting the system, an Intel Virtual RAID on CPU application will be installed onto the system which can be used to manage RAID volumes on the system using
The Userguide provided by Intel on the download site provides steps on setting up different RAID profiles and adjusting settings as needed
In “Configure Volume Name and Size” select the volume name, volume size and strip size for your configuration then click “Next”.
In “Confirm Volume Creation”, you may review the selected configuration, then click “Create Volume”.
Elevated Operating Ambient - If installed in a closed or multi-unit rack assembly, the operating ambient temperature of the rack environment may be greater than room ambient. Therefore, consideration should be given to installing the equipment in an environment compatible with the maximum ambient temperature (Tma) specified by the manufacturer.
Reduced Air Flow - Installation of the equipment in a rack should be such that the amount of air flow required for safe operation of the equipment is not compromised.
Mechanical Loading - Mounting of the equipment in the rack should be such that a hazardous
condition is not achieved due to uneven mechanical loading.
Circuit Overloading - Consideration should be given to the connection of the equipment to the supply circuit and the effect that overloading of the circuits might have on overcurrent protection and supply wiring. Appropriate consideration of equipment nameplate ratings should be used when addressing this concern.
Reliable Earthing - Reliable earthing of rack-mounted equipment should be maintained. Particular attention should be given to supply connections other than direct connections to the branch circuit (e.g. use of power strips).
Ambient operating temperature must be between 5 °C to 40 °C with a non-condensing relative humidity of 8-85%.
The device can be stored at temperatures between -40 °C to 70 °C.
Keep the device away from liquids and flammable materials.
Do not clean the device with liquids. The chassis can be cleaned with a cloth.
Allow at least 2 inches of space around all sides of the device for proper cooling. If the device is mounted to a vertical surface then recommended device orientation is so that heatsink fins allow air to rise unobstructed. Alternative orientations may result in reduced operational temperature range.
This device is intended for indoor operation only.
Install the device only with shielded network cables.
Service and repair of the device must be done by qualified service personnel. This includes, but is not limited to, replacement of the CMOS battery. Replacement CMOS battery must be of the same type as the original.
Proper disposal of CMOS battery must comply with local governance.
Product must only be connected to a certified router, switch or similar network equipment.
Product is intended for indoor use only.
Product cannot be connected to the public network.\
WARNING: There is danger of explosion if the CMOS battery is replaced incorrectly. Disposal of battery into fire or a hot oven, or mechanically crushing or cutting of a battery can result in an explosion.
Température ambiante de fonctionnement élevée - En cas d'installation dans un rack fermé ou à plusieurs unités, la température ambiante de fonctionnement de l'environnement du rack peut être supérieure à la température ambiante de la pièce. Par conséquent, il convient de veiller à installer l'équipement dans un environnement compatible avec la température ambiante maximale (Tma) spécifiée par le fabricant.
Débit d'air réduit - L'installation de l'équipement dans un rack doit être telle que la quantité de débit d'air requise pour un fonctionnement sûr de l'équipement ne soit pas compromise.
Chargement mécanique - Le montage de l'équipement dans le rack doit être tel qu'un condition n'est pas atteinte en raison d'une charge mécanique inégale.
Surcharge de circuit - Il convient de tenir compte de la connexion de l'équipement au circuit d'alimentation et de l'effet que la surcharge des circuits pourrait avoir sur la protection contre les surintensités et le câblage d'alimentation. Une prise en compte appropriée des valeurs nominales de la plaque signalétique de l'équipement doit être utilisée pour répondre à cette préoccupation.
Mise à la terre fiable - Une mise à la terre fiable de l'équipement monté en rack doit être maintenue. Une attention particulière doit être accordée aux raccordements d'alimentation autres que les raccordements directs au circuit de dérivation (par exemple, utilisation de multiprises).
La température ambiante de fonctionnement doit être comprise entre 5 °C et 40 °C avec une humidité relative sans condensation de 8 à 85 %.
L'appareil peut être stocké à des températures comprises entre -40 °C et 70 °C.
Gardez l'appareil à l'écart des liquides et des matériaux inflammables.
Ne nettoyez pas l'appareil avec des liquides. Le châssis peut être nettoyé avec un chiffon.
Laissez au moins 2 pouces d'espace autour de tous les côtés de l'appareil pour un refroidissement correct. Si l'appareil est monté sur une surface verticale, l'orientation recommandée de l'appareil est de sorte que les ailettes du dissipateur thermique permettent à l'air de monter sans obstruction. Des orientations alternatives peuvent entraîner une plage de températures de fonctionnement réduite.
Cet appareil est destiné à une utilisation en intérieur uniquement.
Installez l'appareil uniquement avec des câbles réseau blindés.
L'entretien et la réparation de l'appareil doivent être effectués par un personnel qualifié. Cela inclut, mais sans s'y limiter, le remplacement de la batterie CMOS. La batterie CMOS de remplacement doit être du même type que celle d'origine.
L'élimination appropriée de la batterie CMOS doit être conforme à la gouvernance locale.
Le produit doit uniquement être connecté à un routeur, un commutateur ou un équipement réseau similaire certifié.
Le produit est destiné à une utilisation en intérieur uniquement.
Utilisez uniquement des connecteurs répertoriés UL pour la connexion aux panneaux de fusibles automobiles.
Le produit ne peut pas être connecté au réseau public.
ATTENTION: Il existe un risque d'explosion si la pile CMOS n'est pas remplacée correctement. L'élimination de la batterie dans le feu ou dans un four chaud, ou l'écrasement ou le découpage mécanique d'une batterie peut entraîner une explosion.
Variants
AX300 - Unrestricted Configuration
AX301 - Single CPU, Up to 2x Dual Slot PCIe Adapters + 1x additional Network Adapter - 8x SATA SSD Front Drive Bay
AX302 - Single CPU, Up to 2x Dual Slot PCIe Adapters + 1x additional Network Adapter - 4x NVMe SSD Front Drive Bay
AX303 - Dual CPU, 2x Dual Slot GPUs + 2x Network Adapters with 1x additional Network/GPU/RAID Adapter (Future) - 8x SATA SSD Front Drive Bay
AX304 - Dual CPU, 2x Dual Slot GPUs + 2x Network Adapters with 1x additional Network/GPU/RAID Adapter (Future) - 4x NVMe SSD Front Drive Bay
AX305 - Dual CPU, Up to 7x Single Slot GPUs - 8x SATA SSD Front Drive Bay
AX306 - Dual CPU, Up to 7x Single Slot GPUs - 4x NVMe SSD Front Drive Bay
AX307 - Dual CPU, Up to 4x Dual Slot GPUs - 8x SATA SSD Front Drive Bay
AX308 - Dual CPU, Up to 4x Dual Slot GPUs - 4x NVMe SSD Front Drive Bay
Processor
5th Gen Intel® Xeon® Scalable Processors
Socket: 1+1 Socket E (LGA 4677)
Supported CPUs
Intel 6548Y+ 32 Cores/64 Threads, 205W - 250W TDP¹
Intel 6548N 32 Cores/64 Threads, 205W - 250W TDP¹²
Intel 6538N 32 Cores/64 Threads, 185W - 205W TDP¹²
Intel 6526Y 16 Cores/32 Threads, 150W - 195W TDP¹
Intel 5520+ 28 Cores/56 Threads, 205W TDP
Intel 5515+ 8 Cores/16 Threads, 165W TDP²
Intel 4516Y+ 24 Cores/48 Threads, 150W - 185W TDP¹
Intel 4514Y 16 Cores/32 Threads, 135W - 185W TDP¹
Intel 4510T 12 Cores/24 Threads, 105W - 115W TDP¹
Intel 4510 12 Cores/24 Threads, 150W TDP²
Intel 4509Y 8 Cores/16 Threads, 105W - 125W TDP¹
¹Intel Speed Select Technology (Intel SST-PP) can be used to adjust TDP
²Support for CPU models may be delayed from initial release.
Additional CPU models may be supported based on customer request. Consult OnLogic sales.
Memory
Up to 16x DDR5 RDIMM from 16 to 64 GB in EC8
Maximum operational speed*:
5600 MT/s on 5th Gen Intel Xeon Scalable Processors
*Where higher-speed SKUs do not exist, select 5th Gen Intel Xeon CPUs will use the lower (4800MT/s) speed
FCC 47 CFR Part 15 Subpart B (Class A)
CAN ICES-003(A) / NMB-003(A) (Class A)
EN 63268-1
CISPR 32/EN 55032 (Class A)
CISPR 35/EN 55035
RoHS (2011/65/EU, (EU) 2015/863)
WEEE Directive (2012/19/EU)
Americas
Canada, United States
Europe
Austria, Belgium, Bulgaria, Croatia, Czech Republic, Cyprus, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Liechtenstein, Luxembourg, Malta, Norway, The Netherlands, United Kingdom, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden
Upon Request
Australia, China, Japan, New Zealand
CPU TMAX
GPU TMAX
RAM TMAX
Entry Configuration
84C
89C
56C
Rail SKU
Minimum Span
Maximum Span
Forward Mounting Hole Label
Reverse Mounting Hole Label
Compatible with Cable Management Arm?
3RAMIS214700
430.4mm (16.9”)
600mm (23.6”)
SKU
Description
Filtration
F1-AX301-FILTER
AX300 Front Dust Filter
Debris and large fibers
F1-AX301-FILTER-MERV4
AX300 MERV 4 Front Dust Filter
10.02
07/11/2025
1.03
10/3/2023
1.12.00
04/01/2025
Date
Revision History
01 Dec 2024
First release of Axial AX300 Series Server manual
09 June 2025
Added acoustic & earthquake testing results. Added 28” slide rail.
General information about the BMC, or Baseboard Management Controller, are discussed on our blog post here.
Storage drives are shown in a few different places in the BIOS depending on the type (SATA vs. NVMe) and where its connected (Oculink vs. M.2 PCIe). SATA: Advanced -> Storage Configuration -> SATA_4 – SATA7 visible Oculink: Advanced -> Storage Configuration -> Oculink1_SATA_0 – Oculink1_SATA_3 NVMe: Advanced -> NVME Configuration -> Shows a list of available drives. Select a specific drive to view additional information about it. RAID: Advanced -> Intel® Rapid Storage Technology -> Shows any configurated RAID arrays, and selecting one will display the Selected Disks in the particular RAID volume.
If the system fails to power on or is unresponsive, clearing the CMOS may help. It will also restore the BIOS to factory defaults.
Disconnect the system from all cables/connection (i.e. power, video, etc.) Follow the Opening the System instructions above to gain access to the motherboard. If a PCIe card is installed, you may need to remove it. Follow the Adding/Removing PCIe card instructions above, if needed.
2. Locate the CMOS pads indicated by the orange circle
Once you’ve located the CLRCMOS1 pads, use a screwdriver or other conductive tool to short the pads together for at least 30 seconds.
After at least 30 seconds, the CMOS has been cleared. Reassemble the system and power it back up. The unit may restart several times while the motherboard reinitializes.
Storage drives are shown in a few different places in the BIOS depending on the type (SATA vs. NVMe) and where its connected (Oculink vs. M.2 PCIe). SATA: Advanced -> Storage Configuration -> SATA_4 – SATA7 visible Oculink: Advanced -> Storage Configuration -> Oculink1_SATA_0 – Oculink1_SATA_3 NVMe: Advanced -> NVME Configuration -> Shows a list of available drives. Select a specific drive to view additional information about it. RAID: Advanced -> Intel® Rapid Storage Technology -> Shows any configurated RAID arrays, and selecting one will display the Selected Disks in the particular RAID volume.
Typical Configuration
F2
80% Dust Arrestance (MERV 4)
AX302, AX304, AX306, AX308
Support for up to 4x 2.5" 15mm NVMe Drives in front bay
RAID 0/1/5/10 Support through Intel VROC*
*VROC Standard only supports RAID 0/1/10
