Home
Compute and servers
LinuxONE
IBM LinuxONE is designed to promote sustainable IT by combining energy efficiency with high-performance computing.
By consolidating Linux® workloads into fewer LinuxONE servers, businesses can achieve a 75% reduction in energy consumption, a 50% decrease in floor space usage and lower their CO2e emissions by over 850 metric tons annually2 —directly supporting their environmental goals.
Learn how each LinuxONE model can help you identify opportunities for the greatest greenhouse gas reductions.
What you can do
This guide is a technical roadmap for deploying IBM LinuxONE effectively. It serves as a comprehensive resource for organizations looking to maximize sustainability benefits with LinuxONE, covering best practices for efficient workload consolidation, physical and environmental footprint reduction, and emissions monitoring.
This article explains granular logical partition (LPAR) reporting for precise power consumption insights. Enhanced Hardware Management Console (HMC) Environmental Dashboard and Web Services API capabilities allow IT teams to identify high-energy workloads within specific partitions, optimizing resource use and further progressing toward sustainability goals.
This document explains how to optimize LinuxONE systems for sustainability through the HMC interface. It shares detailed instructions for managing and monitoring LinuxONE systems, including how to use the HMC’s Environmental Dashboard for tracking energy usage and efficiency metrics.
This technical guide describes how to use the HMC Web Services API to access key environmental and power consumption data, including historical data. This allows you to integrate the data into data center infrastructure management (DCIM) systems and other software such as IBM Instana® Observability.
The Power Estimation Tool helps you estimate the power consumption and weight of a specific IBM LinuxONE configuration, to better understand the environmental impact the machine will have to your data center.
Below are other tools to help visualize sustainability metrics for continued progress toward your sustainability goals.
You can integrate HMC data into IBM Instana with the HMC sensor, an optional feature of Instana Observability. You can track your system energy and resource utilization alongside your application performance.
When migrating workloads from x86 to IBM LinuxONE, you might use this Sustainability Grafana Dashboard with Grafana to visualize major sustainability metrics and compare x86 to IBM LinuxONE.
Phoenix Systems used IBM LinuxONE to cut data center floor space by 8x, significantly decreasing energy consumption and maintenance.
Gruppo Boero deployed IBM LinuxONE to drive cost savings through consolidation without sacrificing performance, security, openness or reliability.
Plastic Bank uses IBM LinuxONE in its efforts to clean up the oceans and support coastal communities.
Newlight Technologies uses blockchain technology on IBM LinuxONE to verify a new process that turns greenhouse gas into high-performance biomaterials.
Footnotes
¹ IBM internal tests show that when running WebSphere and DB2 workloads, IBM LinuxONE Emperor 4 requires 16 times fewer cores than the compared x86 servers. If you scale this up to a complete IT solution this means when running this workload, the IBM LinuxONE Emperor 4 Max 125 would be doing the work of about 2000 cores of the compared x86 servers.
DISCLAIMER: This is an IBM internal study designed to replicate a typical IBM customer workload usage in the marketplace. Results might vary. The core consolidation study targeted comparison of the following IBM LinuxONE and x86 servers: IBM LinuxONE Emperor 4 Max 125 system consists of three CPC drawers containing 125 configurable processor units (IFLs) and two I/O drawers to support both network and external storage. Lenovo ThinkSystem SR650 (2U) with two 2nd Gen Intel® Xeon® Platinum processors 2.1 GHz, 16 cores per CPU. Both the x86-based and LinuxONE solutions had access to the same storage array. The workloads consisted of a transactional application running on WebSphere Application Server and IBM DB2 simulating core online banking functions. The actual test results were extrapolated to the stated above x86 servers using IDC QPI metrics and IBM sizing methodology using the following assumptions on a typical IT environment of a banking client using x86 servers. The production IT environment has 16 x86 servers running at 50% average utilization. There are 48 x86 servers in the nonproduction IT environments: development (4 environments with 2 servers each, 8 servers total), development test environment (4 servers), system integration test environment (8 servers), performance test environment (16 servers), user acceptance test environment (4 servers), production fix test environment (8 servers). A typical average CPU utilization is 7% across all nonproduction environments. An equivalent LinuxONE Emperor 4 solution requires a single Max 125 server running at 85% average utilization across all IT environments separated using LPAR technology.
2 Consolidating Linux workloads on 5 IBM LinuxONE Emperor 4 systems instead of running them on compared x86 servers under similar conditions can reduce energy consumption by 75%. It also reduces space by 50%, and the CO2e footprint by over 850 metric tons annually.
DISCLAIMER: Compared 5 IBM Machine Type 3931 Max 125 model consists of three CPC drawers containing 125 configurable cores (CPs, zIIPs or IFLs) and two I/O drawers to support both network and external storage versus 192 x86 systems with a total of 10364 cores. IBM Machine Type 3931 power consumption was based on inputs to the IBM Machine Type 3931 IBM Power Estimation Tool for a memo configuration. x86 power consumption was based on March 2022 IDC QPI power values for 7 Cascade Lake and 5 Ice Lake server models, with 32 to 112 cores per server. All compared x86 servers were 2 or 4 socket servers. IBM Z and x86 are running 24x7x365 with production and non-production workloads. Savings assumes a Power Usage Effectiveness (PUE) ratio of 1.57 to calculate additional power for data center cooling. PUE is based on Uptime Institute 2021 Global Data Center Survey (https://uptimeinstitute.com/about-ui/press-releases/uptime-institute-11th-annual-global-data-center-survey). CO2e and other equivalencies that are based on the EPA GHG calculator (https://www.epa.gov/energy/greenhouse-gas-equivalencies-calculator) use U.S. National weighted averages. Results may vary based on client-specific usage and location.
3 https://cn.wicinternet.org/2023-11/07/content_36950445.htm (link resides outside ibm.com)
4 https://www.opengroup.org/the-open-group-apac-awards-2023 (link resides outside ibm.com)