SPEC CPU2017 Platform Settings for Lenovo Systems

Operating System Tuning Parameters

sched_cfs_bandwidth_slice_us

This OS setting controls the amount of run-time(bandwidth) transferred to a run queue from the task's control group bandwidth pool. Small values allow the global bandwidth to be shared in a fine-grained manner among tasks, larger values reduce transfer overhead. The default value is 5000 (ns).

sched_latency_ns

This OS setting configures targeted preemption latency for CPU bound tasks. The default value is 24000000 (ns).

sched_migration_cost_ns

Amount of time after the last execution that a task is considered to be "cache hot" in migration decisions. A "hot" task is less likely to be migrated to another CPU, so increasing this variable reduces task migrations. The default value is 500000 (ns).

sched_min_granularity_ns

This OS setting controls the minimal preemption granularity for CPU bound tasks. As the number of runnable tasks increases, CFS(Complete Fair Scheduler), the scheduler of the Linux kernel, decreases the timeslices of tasks. If the number of runnable tasks exceeds sched_latency_ns/sched_min_granularity_ns, the timeslice becomes number_of_running_tasks * sched_min_granularity_ns. The default value is 8000000 (ns).

sched_wakeup_granularity_ns

This OS setting controls the wake-up preemption granularity. Increasing this variable reduces wake-up preemption, reducing disturbance of compute bound tasks. Lowering it improves wake-up latency and throughput for latency critical tasks, particularly when a short duty cycle load component must compete with CPU bound components. The default value is 10000000 (ns).

numa_balancing

This OS setting controls automatic NUMA balancing on memory mapping and process placement. NUMA balancing incurs overhead for no benefit on workloads that are already bound to NUMA nodes. Possible settings:

• 0: disables this feature
• 1: enables the feature (this is the default)

For more information see the numa_balancing entry in the Linux sysctl documentation.

kernel.randomize_va_space (ASLR)

This setting can be used to select the type of process address space randomization. Defaults differ based on whether the architecture supports ASLR, whether the kernel was built with the CONFIG_COMPAT_BRK option or not, or the kernel boot options used.
Possible settings:

• 0: Turn process address space randomization off.
• 1: Randomize addresses of mmap base, stack, and VDSO pages.
• 2: Additionally randomize the heap. (This is probably the default.)

Disabling ASLR can make process execution more deterministic and runtimes more consistent. For more information see the randomize_va_space entry in the Linux sysctl documentation.

Transparent Hugepages (THP)

THP is an abstraction layer that automates most aspects of creating, managing, and using huge pages. It is designed to hide much of the complexity in using huge pages from system administrators and developers. Huge pages increase the memory page size from 4 kilobytes to 2 megabytes. This provides significant performance advantages on systems with highly contended resources and large memory workloads. If memory utilization is too high or memory is badly fragmented which prevents hugepages being allocated, the kernel will assign smaller 4k pages instead. Most recent Linux OS releases have THP enabled by default.
THP usage is controlled by the sysfs setting /sys/kernel/mm/transparent_hugepage/enabled. Possible values:

• never: entirely disable THP usage.
• madvise: enable THP usage only inside regions marked MADV_HUGEPAGE using madvise(3).
• always: enable THP usage system-wide. This is the default.

THP creation is controlled by the sysfs setting /sys/kernel/mm/transparent_hugepage/defrag. Possible values:

• never: if no THP are available to satisfy a request, do not attempt to make any.
• defer: an allocation requesting THP when none are available get normal pages while requesting THP creation in the background.
• defer+madvise: acts like "always", but only for allocations in regions marked MADV_HUGEPAGE using madvise(3); for all other regions it's like "defer".
• madvise: acts like "always", but only for allocations in regions marked MADV_HUGEPAGE using madvise(3). This is the default.
• always: an allocation requesting THP when none are available will stall until some are made.

An application that "always" requests THP often can benefit from waiting for an allocation until those huge pages can be assembled.
For more information see the Linux transparent hugepage documentation.

tuned-adm

The tuned-adm tool is a commandline interface for switching between different tuning profiles available to the tuned tuning daemon available in supported Linux distros. The default configuration file is located in /etc/tuned.conf and the supported profiles can be found in /etc/tune-profiles. Some profiles that may be available by default include: default, desktop-powersave, server-powersave, laptop-ac-powersave, laptop-battery-powersave, spindown-disk, throughput-performance, latency-performance, enterprise-storage. To set a profile, one can issue the command "tuned-adm profile (profile_name)". Here are details about relevant profiles:

• throughput-performance: Server profile for typical throughput tuning. This profile disables tuned and ktune power saving features, enables sysctl settings that may improve disk and network IO throughput performance, switches to the deadline scheduler, and sets the CPU governor to performance.
• latency-performance: Server profile for typical latency tuning. This profile disables tuned and ktune power saving features, enables the deadline IO scheduler, and sets the CPU governor to performance.
• enterprise-storage: Server profile to high disk throughput tuning. This profile disables tuned and ktune power saving features, enables the deadline IO scheduler, enables hugepages and disables disk barriers, increases disk readahead values, and sets the CPU governor to performance

dirty_background_ratio

Set through "echo 40 > /proc/sys/vm/dirty_background_ratio". This setting can help Linux disk caching and performance by setting the percentage of system memory that can be filled with dirty pages.

dirty_ratio

Set through "echo 8 > /proc/sys/vm/dirty_ratio". This setting is the absolute maximum amount of system memory that can be filled with dirty pages before everything must get committed to disk.

ksm/sleep_millisecs

Set through "echo 200 > /sys/kernel/mm/ksm/sleep_millisecs". This setting controls how many milliseconds the ksmd (KSM daemon) should sleep before the next scan.

swappiness

The swappiness value can range from 1 to 100. A value of 100 will cause the kernel to swap out inactive processes frequently in favor of file system performance, resulting in large disk cache sizes. A value of 1 tells the kernel to only swap processes to disk if absolutely necessary. This can be set through a command like "echo 1 > /proc/sys/vm/swappiness"

Zone Reclaim Mode

Zone reclaim allows the reclaiming of pages from a zone if the number of free pages falls below a watermark even if other zones still have enough pages available. Reclaiming a page can be more beneficial than taking the performance penalties that are associated with allocating a page on a remote zone, especially for NUMA machines. To tell the kernel to free local node memory rather than grabbing free memory from remote nodes, use a command like "echo 1 > /proc/sys/vm/zone_reclaim_mode"

Free the file system page cache

The command "echo 3> /proc/sys/vm/drop_caches" is used to free pagecache, dentries and inodes.

cpupower

The OS 'cpupower' utility is used to change CPU power governors settings. Available settings are:

• performance: Run the CPU at the maximum frequency.
• powersave: Run the CPU at the minimum frequency.
• ondemand: Scales the frequency dynamically according to current load. Jumps to the highest frequency and then possibly back off as the idle time increases.

Firmware / BIOS / Microcode Settings

Choose Operating Mode: (Default="Maximum Efficiency")

Select the operating mode based on your preference. Note, power savings and performance are also highly dependent on hardware and software running on system.

• "Maximum Efficiency" 'Efficiency' balances performance and power consumption.
• "Maximum Performance"'Performance' maxmizes performance minimizes latency with little regard to power consumption.
• "Custom Mode" When a preset mode is selected, the low-level settings are not changeable and will be grayed out. If user would like to change the settings, please choose Custom Mode. Custom Mode will inherit the UEFI settings from the previous preset operating mode. For example, if the previous operating mode was the Maximum Performance operating mode and then Custom Mode was selected, all the settings from the Maximum Performance operating mode will be inherited.

Determinism Slider:

• "Performance" When set to Performance, performance is more predictable (deterministic) and operates at the lowest common denominator among the cores. But aggregate peak performance may be reduced. Aggregate performance may be higher, but predictability is lower.
• "Power" When set to Power, cores can scale frequency independently.

Core Performance Boost:

Allows the processor to opportunistically increase a set of CPU cores higher than the CPU’s rated base clock speed, based on the number of active cores, power and thermal headroom in a system. When set to Enable, cores can go to turbo frequencies.

4-Link xGMI Max Speed:

This parameter is used to set the maximum xGMI interconnect speed between AMD processors. For NUMA-aware workloads, users can lower the xGMI speed setting to reduce power consumption. Available settings are 18 Gbps, 16 Gbps and 13 Gbps. Default is 13 Gbps.

Global C-state Control:

Controls IO based C-state generation and DF C-states.

cTDP:

Sets the maximum power consumption for CPU. cTDP is only configurable before OS boot.

• "Auto" Sets cTDP=TDP for the installed CPU SKU.
• "Maximum" Maximum sets the maximum allowed cTDP value for the installed CPU SKU.
• "Manual" Usually, maximum is greater than TDP. If a manual value is entered that is larger than the max value allowed, the value will be internally limited to the maximum allowable value.

cTDP Manual:

cTDP is the acronym for Configurable TDP. Some Rome CPU skus support a default TDP and a higher cTDP expressed in Watts. Model Normal TDP Minimum cTDP Maximum cTDP EPYC 7H12 280 225 280 EPYC 7742 225 225 240 EPYC 7702 200 165 200 EPYC 7702P 200 165 200 EPYC 7662 225 225 240 EPYC 7642 225 225 240 EPYC 7502 180 165 200 EPYC 7502P 180 165 200 EPYC 7542 225 225 240 EPYC 7402 180 165 200 EPYC 7402P 180 165 200 EPYC 7302 155 155 180 EPYC 7302P 155 155 180 EPYC 7252 120 120 150 EPYC 7763 280 225 280 EPYC 7713 225 225 240 EPYC 75F3 280 225 280 EPYC 7543 225 225 240 EPYC 7513 200 165 200 EPYC 72F3 180 165 200 EPYC 7313 155 155 180 EPYC 7773X 280 225 280 EPYC 7573X 280 225 280 EPYC 7473X 240 225 280 EPYC 7373X 240 225 280

Memory Speed:

Select the desired memory speed. Faster speeds offer better performance but consume more power.

NUMA Nodes per Socket:

Specifies the number of desired NUMA nodes per socket.

• "NPS0"NPS0 will attempt to interleave the 2 CPU sockets together.
• "NPS1"NPS1 sets one NUMA node per socket.
• "NPS2"NPS2 sets two NUMA nodes per socket, one per Left/Right Half of the SoC.
• "NPS4"NPS4 sets four NUMA nodes per socket, one per Quadrant.

Default is NPS1.

Package Power Limit:

This Parameter sets the CPU package power limit. The maximum value allowed for PPL is the cTDP limit.

• "Auto(default)"Set to maximum value allowed by installed CPU.
• "Manual"Let the user set a power limit and exposes Package Power Limit value applicable for all poplulated processors in the system. If a manual value entered that is larger than the maximum value allowed(cTDP Maximum), the value will be internally limited to maximum allowable value.

SMT Mode:

Can be used to disable symmetric multithreading. To re-enable SMT, a POWER CYCLE is needed after selecting Enable.

ACPI SRAT L3 Cache as NUMA Domain:

When enabled, the last level cache in each CCX in the system will be declared as a separate NUMA domain. It can improve performance for highly NUMA optimized workloads if workloads or components of workloads can be pinned to cores in a CCX and if they can benefit from sharing an L3 cache.

CCD Control:

Sets the number of CCDs to be used. Once this option has been used to remove any CCDs, a POWER CYCLE is required in order for future selections to take effect.

Efficiency Mode:

This setting enables an energy efficient mode of operation internal to AMD EPYC Gen2 processors at the expense of performance. The settings should be enabled when energy efficient operation is desired from the processor.

• "Disable" Use performance optimized CCLX DPM settings
• "Enable(default)"Use power efficiency optimized CCLX DPM settings

LCC as NUMA Node:

Exposes the processor's last level caches as NUMA nodes. When enabled, can improve performance for highly NUMA optimized workloads if workloads or components of workloads can be pinned into the caches.

Zero Output:

When zero output is set to 'Advanced mode' and multiple power supplies are installed in the server, some of the PSUs will be automatically placed into a low power state under light load conditions. This helps to save power

SOC P-states:

When Auto is selected the CPU SOC P-states(uncore P-states) will be dynamically adjusted. That is, their frequency will dynamically change based on the workload. Selecting P0, P1, P2, or P3 forces the SOC to a specific P-state frequency.

L1 Stream HW Prefetcher:

Enable/Disable L1 Stream HW Prefetcher. Fetches the next cache line int to the L1 cache when cached lines are reused within a certain time period or accessed sequentially.

L2 Stream HW Prefetcher:

Enable/Disable L2 Stream HW Prefetcher. Fetches the next cache line int to the L2 cache when cached lines are reused within a certain time period or accessed sequentially.

DRAM Scrub Time:

Memory reliability parameter that sets the period of time between successive DRAM scrub events. Performance may be reduced with more frequent DRAM scrub events.

DLWM Support:

Dynamic Link Width Management allows the processor to reduce the number of active xGMI lanes from 16 to 8 during periods of low socket-to-socket traffic.

Memory interleaving:

This setting allows interleaved memory accesses across multiple memory channels in each socket, providing higher memory bandwidth.

• "Auto(default)"Interleaving is automatically enabled if memory DIMM configuration supports it.
• "Disabled"No interleaving.