Hot topics
Find the latest performance tuning hints and tips and white papers for Linux® on IBM Z® and LinuxONE.
Page navigation
- Red Hat® OpenShift® Container Platform on Z & LinuxONE - Capacity Planning: Five Checkmarks You Don't Want to Miss
- Exploring the performance of network adapters for Linux on IBM Z
- High Availability Clustering with Red Hat Enterprise Linux 8
- Red Hat OpenShift Container Platform on Z - Performance, Quality & Best Practices
- CPUMF - Recommendations for Linux on IBM Z
- High Availability Clustering with RHEL 7/8 and z/VM - Advanced setup
- Monitoring with RHEL 8 and KVM - Getting Started, Hints and Tips
- Elasticsearch on IBM Z - Performance Experiences, Hints and Tips
- High Availability Clustering with RHEL 8 and z/VM - Getting Started, Hints and Tips
- Red Hat OpenShift Container Platform on Z Networking Performance
- Red Hat OpenShift on IBM Z - Performance Experiences, Hints and Tips
- Scaling HyperPAV alias devices on Linux guests on z/VM
- Compiler GCC
- KVM Network Performance - Best Practices and Tuning Recommendations
- KVM Network Performance - Exploitation of IP Routing
- NVMe with LinuxONE: Performance Insights
Red Hat OpenShift Container Platform on Z & LinuxONE - Capacity Planning: Five Checkmarks You Don't Want to Miss
Capacity panning is an essential step for implementing any OpenShift on IBM Z solution to avoid possible pitfalls during the proof-of-concept or production stage. For your estimates, you must factor in resource sharing, virtualization, overcommitment, LPAR weights, entitlements, polarization, and relative shares.
This presentation explores LPAR performance concepts and walks you through z/VM and KVM virtualization best practices - always keeping a sharp focus on OpenShift on IBM Z as the solution platform. The "Five Checkmarks" address some of the most common misconceptions in OpenShift on IBM Z capacity planning.
Creation date
February 2022
Exploring the performance of network adapters for Linux on IBM Z
With OSA-Express and RoCE Express, IBM® provides two types of physical network adapters for Linux standard networking usage on IBM Z. In this presentation, the performance of the most recent OSA-Express, OSA-Express7S 25 GbE, is compared to its predecessor, OSA-Express6S 10 GbE, and additionally to the 25 GbE RoCE Express2.1 on an IBM z15™. It is shown that the OSA-Express7S 25 GbE significantly improves performance compared to its predecessor, while still hitting some limitations. Furthermore, the strengths and weaknesses of the 25 GbE RoCE Express2.1 are highlighted. The presentation concludes describing benefits and downsides of configuring receive packet steering (RPS), a Linux networking feature, with OSA-Express7S and RoCE Express2.1.
Download PDF
Exploring the performance of network adapters for Linux on IBM Z
Creation date
December 2021
High Availability Clustering with Red Hat Enterprise Linux 8
Mission-critical workloads cannot rely on environments with a single point of failure. A high-availability clustering-solution ensures the continuous operation of the environment. Each section provides an overview of the basics followed by an example. Topics include setting up an LPAR cluster with a KVM guest as a resource.
Download PDF
High Availability Clustering with Red Hat Enterprise Linux 8
Creation date
December 2021
Red Hat OpenShift Container Platform on Z - Performance, Quality & Best Practices
Red Hat OpenShift Container Platform as a highly automated hybrid cloud platform supports both traditional and micro-service applications in the cloud. Learn what OCP does to maintain a high service quality and how the quality attributes interact. This presentation examines the OCP quality attributes with a focus on performance when running on IBM Z. Also, find tips for tuning OCP on IBM Z for best performance.
Download PDF
Red Hat
OpenShift Container Platform on Z - Performance, Quality
& Best Practices
Creation date
October 2021
CPUMF - Recommendations for Linux on IBM Z
This presentation gives a brief introduction to the CPU Measurement Facilities. It describes how to enable counter sets and sampling, and how to record and visualize the hardware and software perf events. It also shows how to use perf to record counter statistics and measure the exploitation of specific hardware features through Linux on IBM Z.
Download PDF
CPUMF -
Recommendations for Linux on IBM Z
Latest update
November 2021
High Availability Clustering with RHEL 7/8 and z/VM - Advanced setup
In production environments, applications are often deployed across multiple IBM Z hardware systems to address maintenance windows or unplanned outages. This presentation explores high availability with two Z hardware systems and a z/VM® SSI environment. It examines how high availability can be built with the RHEL HA solution and which challenges arise with two hardware systems.
Download PDF
High Availability Clustering with RHEL 7/8 and z/VM - Advanced setup
Creation date
August 2021
Monitoring with RHEL 8 and KVM - Getting Started, Hints and Tips
Monitoring virtualized environments with numerous KVM guests can be a challenge, and using suitable tooling is critical. This presentation explores the most important KVM related monitoring tools (kvm_stat, sysstat, pcp, and perf_kvm).
Download PDF
Monitoring with RHEL
8 and KVM - Getting Started, Hints and Tips
Creation date
May 2021
Elasticsearch on IBM Z - Performance Experiences, Hints and Tips
Elasticsearch is the most popular enterprise search engine according to db-engines.com. The number of use cases is huge, the community is extremely active, and new releases are made available frequently. Elasticsearch can be embedded into many software stacks, for example Red Hat OpenShift. This presentation summarizes tuning hints and tips that can improve the performance of Elasticsearch significantly when running on Linux on IBM Z.
Download PDF
Elasticsearch on IBM Z - Performance Experiences, Hints and Tips
Creation date
May 2021
High Availability Clustering with RHEL 8 and z/VM - Getting Started, Hints and Tips
Mission-critical workloads must not rely on environments with a single point of failure. A high availability clustering solution ensures the continuous operation of the environment. This presentation provides step-by-step guidance for an active-passive failover setup with Red Hat high-availability on z/VM.
Download PDF
High Availability Clustering with RHEL 8 and z/VM - Getting Started, Hints and Tips
Creation date
March 2021
Red Hat OpenShift Container Platform on Z Networking Performance
Cloud-native applications can use high-performance network connections to achieve high throughput and low latency. This presentation analyzes the performance of the network stack of Red Hat OpenShift on IBM Z in detail and uses this knowledge as a basis for performance tuning of hybrid cloud platforms.
Download PDF
Red Hat
OpenShift Container Platform on Z Networking
Performance
Creation date
November 2020
Red Hat OpenShift on IBM Z - Performance Experiences, Hints and Tips
Red Hat OpenShift enables Hybrid Cloud solutions on the IBM Mainframe. This presentation summarizes the experiences of the Linux on IBM Z performance team when evaluating OpenShift on IBM Z. It also provides hints and tips for improving the performance of an OpenShift cluster for typical cloud-native applications.
Download PDF
Red Hat
OpenShift on IBM
Z - Performance Experiences, Hints and Tips
Creation date
October 2020
Scaling HyperPAV alias devices on Linux guests on z/VM
This white paper analyzes the performance when scaling HyperPAV alias devices in real and virtualized DASD setups. It shows significant performance improvements for different disk I/O workloads on z/VM guests running Linux on IBM Z.
Download PDF
Scaling HyperPAV alias devices on Linux
guests on z/VM
Creation date
April 2020
Compiler GCC
- Development of new GCC features and exploitation of new hardware features for IBM Z
- GCC versions in distributions for IBM Z
- Optimizing C/C++ code
- Special optimization - FDO and LTO
- Influence of compiler versions and compile options on performance
- Options for 31-bit code
Download PDF
Compiler GCC
Creation date
November 2018
Last update
September 2019
KVM Network Performance - Best Practices and Tuning Recommendations
This presentation provides an overview of the different networking configuration choices running KVM guests on the IBM Z platforms. It also provides tuning recommendations for the KVM host and KVM guest environment to achieve greater network performance.
Download PDF
KVM Network Performance - Best Practices and Tuning
Recommendations
Creation date
October 2016
Last update
September 2020
KVM Network Performance - Exploitation of IP Routing
This presentation shows how to set up network connectivity for KVM guests via Open vSwitch on IBM Z platforms using IP routing capabilities of the TCP/IP stack. It shows how to provide a complex network connectivity to a KVM guest including HiperSockets, just defined by software. It also provides a network performance comparison and recommends configurations based on the application scenario.
Download PDF
KVM Network Performance - Exploitation of IP Routing
Creation date
October 2019
NVMe with LinuxONE: Performance Insights
This presentation summarizes a performance study done with NVMe storage that can be used along with IBM LinuxONE Servers. It includes an I/O performance comparison for NVMe storage versus external storage, with various I/O patterns. In addition, NVMe storage failover scenarios and performance states are considered.
Download PDF
NVMe with LinuxONE: Performance Insights
Creation date
October 2019