HPC System Setting

AMD System Setting

systems settings that are required to configure the system for AMD Instinct™ accelerators and to improve the system to obtain optimal performance of the GPUs.

“High Performance Computing (HPC) Tuning Guide for AMD EPYC™ 7002 Series Processors” or “High Performance Computing (HPC) Tuning Guide for AMD EPYC™ 7003 Series Processors”

•        Core C states

•        AMD-PCI-UTIL (on AMD EPYC™ 7002 series processors)

•        IOMMU (if needed)

Recommended settings for the system BIOS

BIOS Setting Location	Parameter	Value	Comments
Advanced   |- PCI Subsystem Settings	Above 4G Decoding	Enabled	GPU Large BAR Support
|- AMD CBS   |- CPU Common Options   |- Performance	Global C-state Control CCD/Core/Thread Enablement   |- SMT Control	Auto Accept Disable	Global Core C-States
|- DF Common Options   |- Memory Addressing	NUMA nodes per socket Memory interleaving	NPS1,2,4 Auto	NUMA Nodes (NPS)
|- Link	4-link xGMI max speed 3-link xGMI max speed	18Gbps 18Gbps	Set AMD CPU xGMI speed to highest rate supported
|- NBIO Common Options	IOMMU PCIe Ten Bit Tag Support Preferred IO Preferred IO Bus Enhanced Preferred IO Mode	Disabled Enable Manual “Use lspci to find pci device id” Enable
|- SMU Common Options	Determinism Control Determinism Slider cTDP Control cTDP Package Power Limit Control Package Power Limit xGMI Link Width Control xGMI Force Link Width xGMI Force Link Width Control APBDIS DF Cstates Fixed SOC Pstate	Manual Power Manual 240 Manual 240 Manual 2 Force 1 Auto P0	Set cTDP to 240W Set Package Power Limit to 240W Set AMD CPU xGMI width to 16 bits
|- UMC Common Options   |- DDR4 Common Options   |-Enforce POR   |- DRAM Controller Configuration   |- DRAM Power Options	|- Overclock |- Memory Clock Speed Power Down Enable	Accept Enabled 1600MHz Disabled	Set to max Memory Speed, if using 3200MHz DIMM’s: 1600MHz. RAM Power Down
|- Security	TSME	Disabled	Memory Encryption

NBIO Link Clock Frequency

The NBIOs (4x per AMD EPYC™ processor) are the serializers/deserializers (also known as “SerDes”) that convert and prepare the I/O signals for the processor’s 128 external I/O interface lanes (32 per NBIO).

LCLK (short for link clock frequency) controls the link speed of the internal bus that connects the NBIO silicon with the data fabric. All data between the processor and its PCIe lanes flow to the data fabric based on these LCLK frequency settings. The link clock frequency of the NBIO components need to be forced to the maximum frequency for optimal PCIe performance.

For AMD EPYC™ 7002 series processors, this setting cannot be modified via configuration options in the server BIOS alone. Instead, the AMD-IOPM-UTIL (see Section 3.2.3) must be run at every server boot to disable Dynamic Power Management for all PCIe Root Complexes and NBIOs within the system and to lock the logic into the highest performance operational mode.

For AMD EPYC™ 7003 series processors, configuring all NBIOs to be in “Enhanced Preferred I/O” mode is sufficient to enable highest link clock frequency for the NBIO components.

Memory Configuration

For the memory addressing modes, especially the number of NUMA nodes per socket/processor (NPS), the recommended setting is to follow the guidance of the “High Performance Computing (HPC) Tuning Guide for AMD EPYC™ 7002 Series Processors” and “High Performance Computing (HPC) Tuning Guide for AMD EPYC™ 7003 Series Processors” to provide the optimal configuration for host side computation.

If the system is set to one NUMA domain per socket/processor (NPS1), bidirectional copy bandwidth between host memory and GPU memory may be slightly higher (up to about 16% more) than with four NUMA domains per socket/processor (NPS4). For memory bandwidth sensitive applications using MPI, NPS4 is recommended. For applications that are not optimized for NUMA locality, NPS1 is the recommended setting.

Operating System Settings

CPU Core State - ‘C States’

There are several Core-States, or C-states that an AMD EPYC CPU can idle within:

•        C0: active. This is the active state while running an application.

•        C1: idle

•        C2: idle and power gated. This is a deeper sleep state and will have a greater latency when moving back to the C0 state, compared to when the CPU is coming out of C1.

Disabling C2 is important for running with a high performance, low-latency network. To disable power-gating on all cores run the following on Linux systems:

$ cpupower idle-set -d 2 

Note that the cpupower tool must be installed, as it is not part of the base packages of most Linux® distributions. The package needed varies with the respective Linux distribution.

For Ubuntu

$ sudo apt install linux-tools-common 

AMD-IOPM-UTIL

This section applies to AMD EPYC™ 7002 processors to optimize advanced Dynamic Power Management (DPM) in the I/O logic (see NBIO description above) for performance. Certain I/O workloads may benefit from disabling this power management. This utility disables DPM for all PCI-e root complexes in the system and locks the logic into the highest performance operational mode.

Disabling I/O DPM will reduce the latency and/or improve the throughput of low-bandwidth messages for PCI-e InfiniBand NICs and GPUs. Other workloads with low-bandwidth bursty PCI-e I/O characteristics may benefit as well if multiple such PCI-e devices are installed in the system.

The actions of the utility do not persist across reboots. There is no need to change any existing firmware settings when using this utility. The “Preferred I/O” and “Enhanced Preferred I/O” settings should remain unchanged at enabled.

The recommended method to use the utility is either to create a system start-up script, for example, a one-shot systemd service unit, or run the utility when starting up a job scheduler on the system. The installer packages (see Power Management Utility at https://developer.amd.com/iopm-utility/) will create and enable a systemd service unit for you. This service unit is configured to run in one-shot mode. This means that even when the service unit runs as expected, the status of the service unit will show inactive. This is the expected behavior when the utility runs normally. If the service unit shows failed, the utility did not run as expected. The output in either case can be shown with the systemctl status command.

Stopping the service unit has no effect since the utility does not leave anything running. To undo the effects of the utility, disable the service unit with the systemctl disable command and reboot the system.

The utility does not have any command-line options, and it must be run with super-user permissions.

Systems with 256 CPU Threads - IOMMU Configuration

For systems that have 256 logical CPU cores or more (e.g., 64-core AMD EPYC™ 7763 in a dual-socket configuration and SMT enabled), setting the IOMMU configuration to “disabled” can limit the number of available logical cores to 255. The reason is that the Linux® kernel disables X2APIC in this case and falls back to APIC, which can only enumerate a maximum of 255 (logical) cores.

If SMT is enabled by setting “CCD/Core/Thread Enablement > SMT Control” to “enable”, the following steps can be applied to the system to enable all (logical) cores of the system:

•        In the server BIOS, set IOMMU to “Enabled”.

•        When configuring the Grub boot loader, add the following arguments for the Linux kernel:

amd_iommu=on iommu=pt

•        Update Grub to use the modified configuration:

sudo grub2-mkconfig -o /boot/grub2/grub.cfg

•        Reboot the system.

•        Verify IOMMU passthrough mode by inspecting the kernel log via dmesg: