SPEC CPU2006 Platform Settings for Intel-based systems

Firmware / BIOS / Microcode Settings

Hardware Prefetcher:

This BIOS option allows the enabling/disabling of a processor mechanism to prefetch data into the cache according to a pattern-recognition algorithm In some cases, setting this option to Disabled may improve performance. Users should only disable this option after performing application benchmarking to verify improved performance in their environment.

Adjacent Cache Line Prefetch:

This BIOS option allows the enabling/disabling of a processor mechanism to fetch the adjacent cache line within a 128-byte sector that contains the data needed due to a cache line miss. In some cases, setting this option to Disabled may improve performance. Users should only disable this option after performing application benchmarking to verify improved performance in their environment.

High Bandwidth:

Enabling this option allows the chipset to defer memory transactions and process them out of order for optimal performance.

Data Reuse:

Enabling this BIOS option reduces the frequency of L3 cache updates from L1. This may improve performance by reducing the internal bandwidth consumed by constantly updating L1 cache lines in L3. Since this results in more fetches to main memory, setting this option to Disabled may improve performance in some cases. Users should only disable this option after performing application benchmarking to verify improved performance in their environment.

Logical Processor:

This BIOS setting enables/disables Intel's Hyper-Threading (HT) Technology. With HT Technology, the operating system can execute two threads in parallel within each processor core.

Execute disable:

This is a security feature designed to prevent certain types of buffer overflow attacks by enforcing specific areas of memory that applications can execute code. In general, it is best to leave this option Enabled for the security benefits, as no real performance advantage has been detected by disabling this feature in BIOS

Virtualization technology:

When this option is enabled, the BIOS will enable processor Virtualization features and provide the virtualization support to the OS through the DMAR table. In general, only virtualized environments such as VMware® ESX™, Microsoft® Hyper-V™ , Red Hat® KVM, and other virtualized operating systems will take advantage of these features. Disabling this feature is not known to significantly alter the performance or power characteristics of the system, so leaving this option Enabled is advised for most cases.

Node Interleaving:

This BIOS option allows the enabling/disabling of memory interleaving across CPU nodes. When disabled, each CPU chip can only access memory within its own node.

CPU Power Management:

This BIOS setting allows configuration of various demand-based switching schemes. Maximum Performance maintains full voltage to processor internal components, even during periods of inactivity, eliminating the performance penalty associated with the phase transitions between high and low load.

Memory Frequency:

This BIOS setting allows the memory to be clocked to the highest supported frequency.

Turbo Boost:

Intel Turbo Boost Technology is a processor feature which allows the processor to transition to a higher frequency than the processor's rate speed if the processor has available power headroom and is within temperature specifications. Disabling this feature will reduce power usage but will reduce the system's maximum achievable performance under some workloads.

C States:

Enabling the CPU States causes the CPU to enter a low-power mode when the CPU is idle.

System Profile:

This BIOS option sets the performance and power management aggressiveness for the system. It is a collection of selections including a custom selection designed to allow customers to choose the ideal operating profile for their server system environment. It includes settings like CPU Power Management, Memory Frequency, Turbo Boost, C1E and C States. Performance profile provides potentially increased performance by maximizing processor frequency and the disabling certain power saving features such as C-states and. Although not optimal for all environments, this is an excellent starting point for performance optimization baseline comparisons.

Memory Patrol Scrub:

Memory Patrol Scrub Patrol Scrubbing is a custom System Profile option feature that scans the memory for bit errors and corrects them whenever possible. When set to Disabled, no patrol scrubbing will occur. When set to Standard Mode, the entire memory array will be scrubbed once in a 24 hour period. When set to Extended Mode, the entire memory array will be scrubbed every hour to further increase system reliability.

Snoop Mode:

Selects between three processor cache coherency snoop modes: Cluster on Die, Home Snoop, Early Snoop. There are 3 snoop mode options for how to maintain cache coherency across the Intel QPI fabric, each with varying memory latency & bandwidth characteristics depending on how the snoop traffic is generated.

Cluster on Die mode logically splits a socket into 2 NUMA domains that are exposed to the OS with half the amount of cores and LLC assigned to each NUMA domain in a socket. This mode utilizes an on-die directory cache & in memory directory bits to determine whether a snoop needs to be sent. Use this mode for highly NUMA optimized workloads to get the lowest local memory latency and highest local memory bandwidth for NUMA workloads.

In Home Snoop and Early Snoop modes, snoops are always sent, they just originate from different places: the caching agent (earlier) in Early Snoop mode and the home agent (later) in Home Snoop mode. Use Home Snoop mode for NUMA workloads that are memoy bandwidth sensitive and need both local & remote memory bandwidth. Use Early Snoop mode for workloads that are memory latency sensitive or for workloads that benefit from fast cache-to-cache transfer latencies from the remote socket. Snoops are sent out earlier, which is why memory latency is lower in this mode.