The IBM® Power11 server family is here. For more on the overall announcement, read the full press release here. Power11 has built upon the long lineage of Power® processor-based technology, trusted by clients around the globe to run business-critical applications and workloads in regulated industries such as healthcare, banking, financial services, telecommunications, government, and more. Now with the AI era fully upon us, we redesigned Power11 up and down the entire stack to deliver always-on IT with consistency across the hybrid cloud to help our clients maintain competitiveness in this new era. With Power11, clients can expect new innovations in three main pillars:
Ensure Always-On IT
The Power11 family provides clients with a new solution and capabilities to ensure always-on IT in industries where downtime is not an option. IT teams often spend a lot of time managing maintenance, including planning for windows of downtime that align with the priorities of the business. With Power11 processor-based servers, clients can expect zero planned downtime for system maintenance through new technologies such as autonomous patching and the automated movement of workloads with Live Partition Mobility[1]. IBM Power Cyber Vault, an integrated cyber resiliency solution consisting of Power11 processor-based servers, IBM FlashSystem® or DS8000® storage, software, and services from IBM Technology Expert Labs, can detect and respond to ransomware threats in less than a minute, enabling protection against cyberattacks and preventing financial damage to the business[2]. Learn more about IBM Power Cyber Vault in this blog.
Boost Outputs and Save Time
Power11 introduces the concept of Resource Groups in the firmware to maximize system utilization and increase performance of workloads efficiently. To further maximize efficiency, Power11 processor-based servers have a new Energy Efficient mode. This new smart energy option automatically schedules workloads to run at the lowest wattage possible to decrease energy costs and reduce environmental impact, helping meet corporate and federal sustainability objectives. Using the new Energy Efficient mode, clients can get up to 28% better energy efficiency on Power11 processor-based servers compared to Maximum Performance mode[3]. Power11 processor-based servers also have automated data collection capabilities that send data directly to IBM Technology Lifecycle Services (TLS) for faster error resolution.
Flex Seamlessly to Business Needs
The Power11 portfolio provides the flexibility clients need to scale up and down to meet new business demands and opportunities where they are. Power11 processor-based servers will be available in the public cloud on Power Virtual Server the same day as the on-premise hardware, ensuring a consistent hybrid cloud user experience across public and private environments. The ecosystem of Red Hat and open source solutions available on Power is expanding to build new applications and modernize existing ones. Power11 processor-based servers are AI-era ready with on-chip AI acceleration available today to run large and small language models on the same system that data is living, reducing latency and increasing inferencing performance without the added cost and complexity of GPUs. With the Power S1122, clients can reduce latency by up to 58% for AI inferencing compared to the Power S1022[4].
IBM Power E1180
The IBM Power E1180 is a full-rack server engineered for the largest and most complex business-critical workloads. It harnesses up to 256 Power11 cores and 64 TB of DDR5 memory, and runs IBM® AIX®, IBM i, and Linux. The Power E1180 delivers continuous availability, advanced cyber resilience, and enterprise-grade AI acceleration. The scalability of the Power E1180 helps streamline operations and manage growing data demands with confidence.
The Power E1180 has 10% more rPerf per watt compared to the Power E1080 for increased performance of data-intensive workloads such as SAP HANA[5]. Using the new Energy Efficient mode compared to Maximum Performance mode on Power E1180, clients can get 22% better server efficiency, ideal for reducing the environmental footprint of large enterprise data centers[6]. By leveraging Resource Groups for shared processor configurations on Power E1180, clients can further increase efficiency with up to 25% improvement in per core performance[7]. Learn more about the Power E1180 and read the data sheet here.
IBM Power E1150
The IBM Power E1150 is a 4U server built for memory- and data-intensive workloads in midsize companies to the largest enterprises in the world. It runs AIX and Linux and has up to 120 Power11 cores and 16 TB of DDR5 memory to support multiple business-critical workloads in hybrid cloud environments. The Power E1150 has a space-conscious design that helps reduce costs, simplify operations, and scale workloads without over-provisioning, making it ideal for organizations modernizing infrastructure while focusing on ROI and agility.
As the Power E1150 is a popular choice for running transactional database workloads, improving both performance and efficiency for Power11 was critical. The Power E1150 delivers a 55% improvement in per core database performance with a 25% better transactional response time compared to the Power E950[8]. Compared to Power10, the E1150 has 20% more rPerf per watt over the E1050 to further increase performance of these database workloads while maintaining efficiency[9]. Using the new Energy Efficient mode on the Power E1150, clients will also see 28% better energy efficiency compared to Maximum Performance mode[10]. Learn more about the Power E1150 and read the data sheet here.
IBM Power S1124
The IBM Power S1124 is a 4U server designed for enterprises and regional data centers that need strong compute, memory, and hybrid cloud flexibility. It runs AIX, IBM i, and Linux, and has up to 60 Power11 cores and 8 TB of DDR5 memory to power a variety of business-critical workloads. The Power S1124 enables solid business continuity, seamless scalability, operational efficiency, and modern workload management without adding IT complexity for organizations.
The Power11 processor-based scale-out portfolio has no shortage of performance and efficiency improvements either. Energy Efficient mode is also available on the Power S1124 and Power S1122, and on the Power S1124 clients will see an improvement in energy efficiency of 26% compared to Maximum Performance mode[11]. The Power S1124 also has 22% more rPerf per watt compared to the Power S1024 to further increase performance of business-critical workloads such as IBM i-based applications, Oracle databases, and more[12]. Learn more about the Power S1124 and read the data sheet here.
IBM Power S1122
The IBM Power S1122 is a dense 2U rack server designed for small to midsized businesses, remote branches, and space-constrained environments. It runs AIX, IBM i, and Linux, and has up to 60 Power11 cores and 4 TB of DDR5 memory. The Power S1122 is built for cost-efficiency, hybrid cloud flexibility, and simplified management. It helps organizations run mixed workloads, scale easily, and modernize infrastructure without complex replatforming or rewriting of business-critical applications.
The Power S1122, along with the Power E1180, will also be available as an option on Power Virtual Server for a consistent user experience across on-premise and public cloud. Energy Efficient mode is also available on the Power S1122, with 13% better energy efficiency compared to Maximum Performance mode[13]. The Power S1122 has up to 20 more Power11 cores available on the system compared to a similar Power10 processor-based S1022. This leads to better system utilization and performance, with 37% more rPerf per watt on the Power S1122 compared to the Power S1022[14]. For AI inferencing, clients can get up to 6.9X better performance per dollar with the Power S1122 compared to the Power S1022[15]. Learn more about the Power S1122 and read the data sheet here.
IBM Power Virtual Server
IBM Power Virtual Server is a cloud-based infrastructure solution that brings the performance and reliability of Power to a flexible and scalable virtual environment. It allows businesses to run AIX, IBM i, and Linux workloads in the cloud without having to refactor applications, making it ideal for hybrid cloud strategies and architectures. With seamless integration into the broader IBM Cloud® ecosystem, Power Virtual Server provides secure, high-performance compute capacity alongside automation, backup, and disaster recovery capabilities. It is designed for enterprises looking to modernize their infrastructure, improve operational efficiency, and extend existing on-premise Power environments into the cloud to take advantage of pay-as-you-use billing to manage costs.
For the first time ever, Power11 processor-based servers will be available in the cloud on Power Virtual Server at the same time as General Availability of the on-premise hardware. Power Virtual Server is also now certified as an SAP RISE Hyperscaler on IBM Cloud. With Power11 processor-based servers now on the cloud in Power Virtual Server, performance and agility take another leap forward. And because IBM fully manages the infrastructure, clients can simplify operations and focus on driving outcomes rather than maintaining hardware. Learn more about Power Virtual Server here.
GA is July 25, 2025
The Power E1180, Power E1150, Power S1124, and Power S1122 will be Generally Available on Friday, July 25th, as well as Power11 processor-based servers in the cloud on Power Virtual Server. Keep an eye on the portfolio even after General Availability, as continuous new updates will be rolled out throughout the rest of the year and beyond. As we continue on our journey into the AI era, the team will make IBM® watsonx.data® available on Power by the end of the year to help our clients access, transform, and manage sensitive enterprise data at scale for analytics. The IBM Spyre™ Accelerator for off-chip AI acceleration is expected to be available later this year for Power as well. IBM watsonx Code Assistant™ for i will also be available by the end of the year to help developers modernize business-critical RPG applications with greater ease and productivity. To learn more about Power11 and these announcements, register for the virtual event here today if you haven’t already or catch the replay later on.
Footnotes
[1] Based upon IBM internal testing of system upgrade scenarios; many (i.e. VIOS, hot plug adapters, I/O adapter FW, and concurrent system firmware updates) can be done in-place while some (i.e. non-concurrent system FW and HW maintenance) may require Live Partition Mobility (LPM) support.
[2] This guarantee covers only the displaying of an alert in less then one minute. Remediation is in the form of drive replacement up to the cost of the Covered Product. Terms and conditions apply, full details can be found here.
[3] Based upon IBM measurements of performance per watt on servers comparing Maximum Performance Mode to Energy-Efficient Mode while running compute-, disk-, and memory-based workloads on Power11 systems with fully configured sockets and memory as follows: E1180 with 4x10c / 64x64GB DDIMM, E1150 with 4x16c / 64x32GB DDIMM , S1124 with 2x16c / 32x32GB DDIMM , S1122 with 2x16c / 32x32GB DDIMM
[4] Based on IBM internal testing of batched processing and measuring time (minutes) per documents for Red Hat OpenShift Container Platform 4.16 worker nodes running Summarization with Granite3.3-2b-instruct input ~560 tokens, output 200 tokens, batch size=1. Results valid as of June 23, 2025 and conducted under laboratory conditions, individual results can vary based on workload size, use of storage subsystems and other conditions. Comparison is based on an IBM Power S1122 (2x30-core) with 48 worker cores and S1022 (2x20-core) with 32 worker cores.
[5] Based on maximum configuration at 100% utilization under typical operating conditions where Power E1080 is 7999 rPerf @ 17,392 Watts (0.460 rPerf/Watt), Power E1180 is 8875E rPerf @ 18,048E Watts (0.492 rPerf/Watt); 0.492 / 0.460 = .1.07 more rPerf/Watt
[6] Based upon IBM measurements on servers comparing Maximum Performance Mode to Energy-Efficient Mode while running compute-, disk-, and memory-based workloads running on Power E1180 with 4x10c / 64x64GB DDIMM
[7] Based on IBM internal measurements of a commercial transactional database(OLTP) workload consisting of 10 partitions executing on an E1080 (16x12 Power10 CPUs @ 3.6-4.15GHz) compared to the same workload executing on an E1180 (16x12 Power11 CPUs @ 3.9-4.4GHz) within 2 Resource Groups with 4 and 6 partitions.
[8] Based on IBM internal measurements for similar throughput of a commercial banking transactional database(OLTP) workload executing on 1xE1150 with 9 Power11 CPUs @ 3.5-4.2GHz compared 1xE950 with 14 Power9 CPUs @ 3.2-3.8GHz
[9] Based on maximum configuration at 100% utilization under typical operating conditions where Power E1050 is 2689 rPerf @ 3,840 Watts (0.700 rPerf/Watt), Power E1150 is 3402E rPerf @ 4,168E Watts (0.816 rPerf/Watt); 0.816 / 0.700 = .1.17 more rPerf/Watt
[10] Based upon IBM measurements on servers comparing Maximum Performance Mode to Energy-Efficient Mode while running compute-, disk-, and memory-based workloads running on Power E1150 with 4x16c / 64x32GB DDIMM
[11] Based upon IBM measurements on servers comparing Maximum Performance Mode to Energy-Efficient Mode while running compute-, disk-, and memory-based workloads running on Power S1124 with 2x16c / 32x32GB DDIMM
[12] Based on maximum configuration at 100% utilization under typical operating conditions where Power S1024 is 1331 rPerf @ 2,707 Watts (0.492 rPerf/Watt), Power S1124 is 1686E rPerf @ 2,907E Watts (0.580 rPerf/Watt); 0.580 / 0.492 = .1.18 more rPerf/Watt
[13] Based upon IBM measurements on servers comparing Maximum Performance Mode to Energy-Efficient Mode while running compute-, disk-, and memory-based workloads running on Power S1122 with 2x16c / 32x32GB DDIMM
[14] Based on maximum configuration at 100% utilization under typical operating conditions where Power S1022 is 1024 rPerf @ 1,684 Watts (0.608 rPerf/Watt), Power S1124 is 1487E rPerf @ 1,834E Watts (0.811 rPerf/Watt); 0.811 / 0.608 = .1.33 more rPerf/Watt
[15] Based on IBM internal testing of batched processing and measuring documents per hour for Red Hat OpenShift Container Platform 4.16 worker nodes running Summarization with Granite3.3-2b-instruct input ~560 tokens, output 200 tokens, batch size=8. Results valid as of June 23, 2025 and conducted under laboratory conditions, individual results can vary based on workload size, use of storage subsystems and other conditions. Comparison is based on an IBM Power S1122 (2x30-core) with 48 worker cores and S1022 (2x20-core) with 32 worker cores.