Skip to main content
All CollectionsDecember 31, 2023PortalGeneral
What is a Compute Unit (CU) in OneIQ
What is a Compute Unit (CU) in OneIQ

Understanding what a Compute Unit is and where it's used.

Phi Team avatar
Written by Phi Team
Updated over 3 weeks ago

Overview

A Compute Unit (CU) in OneIQ is a standardized measure of CPU performance used mainly for sizing calculations. Traditional metrics like GHz are not an accurate measure of computing power, as older CPUs with higher clock speeds may perform worse than modern processors with lower GHz ratings. CUs address this by normalizing performance across different CPUs using a common benchmark, factoring in both processing capability and speed. This ensures a consistent and reliable way to compare compute capacity for infrastructure planning and optimization.

Key Details

Definition of Compute Unit (CU)

A Compute Unit (CU) is a measure used by OneIQ to standardize CPU performance across various CPUs.

We assign a performance rating to each CPU and use this rating to compare and normalize CPU values against the Intel Xeon Processor E5-2609 v2 (10M Cache, 2.50 GHz), establishing a common benchmark.

1 Compute Unit is equivalent to a fully utilized single core of an Intel Xeon Processor E5-2609 v2 (10M Cache, 2.50 GHz).

Purpose of Compute Units (CUs)

Standardization: CUs provide a common baseline to compare different CPUs, making it easier to assess the overall compute capacity of mixed environments.

Simplified Display: By converting diverse CPU speeds into CUs, OneIQ presents a clear, comparable view of computing power.

How CUs Work in OneIQ

Normalization: Servers in an IT environment may have various processor models, each with different performance capabilities.

Benchmarking: OneIQ uses CPU benchmarks to measure and normalize CPU performance, converting it into CUs. This allows all CPU data to be displayed in a unified format, regardless of the underlying hardware differences.

Practical Use of CUs in OneIQ

A CU is primarily an internal metric used within One IQ for Right-Sizing Solutions. CUs help determine the appropriate computing resources needed for specific workloads, allowing for precise infrastructure design and optimization.

Did this answer your question?