Power

 General description

PowerVM provides a secure and scalable virtualization environment for AIX, IBM i and Linux applications built upon the advanced RAS features and leading performance of the Power Systems platform. IBM PowerVM provides the industrials trength virtualization solution for IBM Power Systems servers and blades. IBM Power Systems servers integrated with PowerVM technology are designed to allow clients to build a dynamic infrastructure that will help them to reduce costs, manage risk and improve service levels. The POWER Hypervisor is a firmware layer sitting between the hosted operating systems and the server hardware, as shown in Figure below:
Fig. 4.6.3/1: Power Hypervisor abstracts physical server hardware.
Fig. 4.6.3/1: Power Hypervisor abstracts physical server hardware.
The Hardware Management Console (HMC) provides functions to simplify the handling of the Virtual I/O Server environment. For example, an overview of the Virtual Ethernet and Virtual SCSI topologies is available. It is also possible to execute Virtual I/O Server commands from the HMC.
Fig. 4.6.3/2: HMC window to activate PowerVM feature.
Fig. 4.6.3/2: HMC window to activate PowerVM feature.
The Integrated Virtualization Manager (IVM) is used to manage selected Power Systems servers using a Web-based graphical interface without requiring an HMC. This reduces the hardware needed for the adoption of virtualization technology, particularly for lowend systems. This solution fits in small and functionally simple environments where only few servers are deployed or some advanced HMC-like functions are required. The Integrated Virtualization Manager (IVM) is a basic hardware management solution, included in the VIO software that inherits key Hardware Management Console (HMC) features.
When using the IVM within the Virtual I/O Server to manage a single system. The Advanced System Management Interface (ASMI) menu to enable the Virtualization Engine Technologies.

 IBM PowerVM features and technologies

The following features and technologies are key to IBM PowerVM:
  • Processor virtualization - The Power Systems family gives the freedom to run the broadest selection of enterprise applications without the costs and complexity often associated with managing multiple physical servers. PowerVM can help eliminate underutilized servers because it is designed to pool resources and optimize their use across multiple application environments and operating systems. Through advanced virtual machine (VM) capabilities, a single VM can act as a completely separate AIX, IBM i, or Linux operating environment, using dedicated or shared system resources. With shared resources, PowerVM can automatically adjust pooled processor, memory or storage resources across multiple operating systems, borrowing capacity from idle VMs to handle high resource demands from other workloads. With PowerVM on Power Systems, you have the power and flexibility to address multiple system requirements in a single machine. PowerVM MicroPartitioning supports multiple VMs per processor core and, depending upon the Power Systems model, can run up to 10003 VMs on a single server—each with its own processor, memory, and I/O resources. Processor resources can be assigned at a granularity of 1/100th of core. Enables up to 20 VMs per processor core. Consolidating systems with PowerVM can help cut operational costs, improve availability, ease management and improve service levels, while allowing businesses to quickly deploy applications. Multiple Shared Processor Pools allows for the automatic non-disruptive balancing of processing power between VMs assigned to shared pools, resulting in increased throughput. It also provides the ability to cap the processor core resources used by a group of VMs to potentially reduce processor-based software licensing costs. Shared Dedicated Capacity allows for the “donation” of spare CPU cycles from dedicated processor VMs to a Shared Processor Pool. Since a dedicated VM maintains absolute priority for CPU cycles, enabling this feature can increase system utilization without compromising the computing power for critical workloads. Because its core technology is built into the system firmware, PowerVM offers a highly-secure virtualization platform that has received the Common Criteria Evaluation and Validation Scheme (CCEVS) EAL4+ certification4 for its security capabilities. The sharing mode of processor cores can be set to capped or uncapped. Capped partitions have a preset amount of maximum processing unit entitlement. However, partitions that are configured with uncapped processor resources are able to use all of their allocation, plus any unused processing units in a shared processor pool. The load on the Virtual I/O Server varies depending on the demands of the virtual clients. Therefore, you might see spikes in processor or memory usage throughout the day. To address these changes in workload, and to achieve better utilization, use shared and uncapped processors. Therefore, uncapping can provide a significant benefit to partitions that have spikes in utilization.
  • Memory virtualization - PowerVM features Active Memory™ Sharing, a technology that allows you to intelligently and dynamically reallocate memory from one VM to another for increased utilization, flexibility and performance. Active Memory Sharing enables the sharing of a pool of physical memory among VMs on a server, helping to increase memory utilization and drive down system costs. The memory is dynamically allocated amongst the VMs as needed, to optimize the overall physical memory usage in the pool.Active Memory Deduplication is a powerful optimization feature that can be enabled when Active Memory Sharing is in being used. This memory optimization intelligently detects and removes duplicate memory pages used between VMs and as a result reduces overall memory consumption.
  • I/O virtualization - The Virtual I/O Server (VIOS) is a special purpose VM that can be used to virtualize I/O resources for AIX, i, and Linux client VMs. VIOS owns the resources that are shared with clients. A physical adapter assigned to the VIOS can be shared by one or more other VMs. VIOS is designed to reduce costs by eliminating the need for dedicated network adapters, disk adapters and disk drives, and tape adapters and tape drives in each client VM. With VIOS, client VMs can easily be created for test, development, or production purposes.Shared Storage Pools allow storage subsystems to be combined into a common pool of virtualized storage that can be shared by the VIOS on multiple Power Systems servers. Shared storage pools support capabilities such as thin provisioning, whereby VM storage is dynamically allocated and released as required, to improve overall storage resource utilization.N_Port ID Virtualization (NPIV) provides direct access to Fiber Channel Adapters from multiple VMs, simplifying the deployment and management of Fiber Channel SAN environments.
Fig. 4.6.3/3: Simple Virtual I/O Server configuration.
Fig. 4.6.3/3: Simple Virtual I/O Server configuration.
  • Live Partition Mobility - Live Partition Mobility supports the movement of a running AIX or Linux or IBM i VM from one Power Systems server to another without application downtime, helping avoid application interruption for planned system maintenance, provisioning, and workload management. Live Partition Mobility can be used to simplify migration of operating environments to new servers temporarily or permanently.
Fig. 4.6.3/4: An Example for Live Partition Mobility.
Fig. 4.6.3/4: An Example for Live Partition Mobility.
  • Real- time Performance Monitoring - Understanding the performance of virtualized workloads is key to optimizing systems virtualized with PowerVM. PowerVM Enterprise Edition includes as a feature IBM PowerVP™ which is a performance monitoring solution that shows how virtual machines map to physical hardware components. PowerVP provides a color coded performance health view using a customizable set of thresholds. This monitor shows potential bottlenecks and helps with optimal VM placement within a virtualized server.
  • Systems management - PowerVM virtualization features are managed through the Hardware Management Console (HMC) or the Flex System Manager, which is part of the IBM PureFlex System, or by the Integrated Virtualization Manager (IVM) on entry-level Power Systems.The Flex System Manager provides virtualization management and system management for the PureFlex System. IVM allows you to point click and consolidate workloads with its easy-to- use web- based interface. IVM lowers the cost of entry into POWER7 processorbased virtualization since it does not require the use of a centralized management console for system management. With IVM, you can manage a single system, including the creation of VMs, virtualized storage and virtualized networking. IBM Systems Director VMControl™ also supports the PowerVM environment. VMControl is the IBM virtualization management tool for multiple, heterogeneous servers. With VMControl you can manage PowerVM for your Power virtualization from the same screen as VMWare for your x86 servers such as IBM System x and BladeCenter. VMControl is a plug-in for IBM Systems Director that supports advanced management functions such as health check and topology mappings, as well as the ability to take action on monitored events. VMControl simplifies the creation and management of standardized virtual appliances (ready-to- run VMs) and system pools-combinations of VMs on multiple servers that can be managed as a single entity.IBM PowerVC™ Virtualization Center is an advanced virtualization management solution which is completely focused on Power Systems virtualization management and is powered by OpenStack. PowerVC provides a very simple virtualization management user experience which manages the complete life cycle of virtual machines from initial image capture to deployment and movement for optimization through decommission-ing of VMs. PowerVC is the next generation of advanced systems management solution for Power Systems.
  • Diverse set of workloads - many IBM Software Group offerings are optimized for a PowerVM environment supporting the consolidation of a diverse set of workloads — from database and application servers to web infrastructure. For example, PowerVM and WebSphere® Virtual Enterprise work together to provide a cross system virtualized application infrastructure that can lower operational and energy costs required to create, run, and manage enterprise applications and SOA environments. WebSphere Virtual Enterprise increases flexibility and agility to ensure business process integrity, improve service and applica-tion performance, and better manage application health.
  • Availability - Many customers want to set up highly available storage environments. Power VM allows them to configure the same using Virtual SCSI and Virtual Fibre Channel by having redundancy. Redundancy of Virtual SCSI and Virtual Fibre Channel can be achieved using MPIO and LVM mirroring in the client partition and also in Virtual I/O Server. One possible configurration if shown on the figure- in the Power VM with Dual Virtual I/O Server environment where the client partition has two virtual SCSI client adapters and each of them is mapped to two different Virtual SCSI server adapters on different Virtual I/O Servers. Each Virtual I/O Server maps the same physical volume to the Virtual SCSI Server adapter on them. The client partition sees the same hdisk, mapped from two Virtual I/O Servers using Virtual SCSI. To achieve this, the same storage needs to be zoned to Virtual I/O Servers from the storage subsystem. This configuration also has redundancy at the Virtual I/O Server physical Fibre Channel adapter.
Fig. 4.6.3/5: Redundancy of Virtual SCSI using Dual Virtual I/O Server
Fig. 4.6.3/5: Redundancy of Virtual SCSI using Dual Virtual I/O Server
  • PowerVM with PowerHA – PowerHA is the IBM important high availability solution for UNIX (AIX) and IBM I systems. PowerHA can work with PowerVM to achieve higher availability.
Fig. 4.6.3/6: An Example for Live Partition Mobility.
Fig. 4.6.3/6: An Example for Live Partition Mobility.
In this case, you can use Live Partition Mobility or PowerHA to migrate or move the application from one system to another to avoid any planned outage. On the other hand, PowerHA monitors the running status for all the nodes of the cluster; in our case, the nodes PowerHA_node 1 and PowerHA_node 2. If one node failed due to some reason, PowerHA can realize it and trigger a series of actions, and bring the workload up on another available node based on the configured policy. That helps you to reduce the unplanned outage of the system.

 Things to consider before choosing PowerVM

Consider the following pros and cons before deciding to use PowerVM as your virtualization tool.
On the pro side
  • PowerVM supports multiple operating environments on a single system.
  • Enables up to 10 VMs per processor core.
  • Processor, memory, and I/O resources can be dynamically moved between VMs.
  • VMs can use dedicated or shared (capped or uncapped) processor resources.
  • Processor resources can automatically move between VMs based on workload demands.
  • Processor resources for a group of VMs can be capped, reducing software license costs.
  • Storage resources for Power Systems servers and VIOS can be centralized in pools to optimize resource utilization.
  • Simplifies VM creation and management for entry Power Systems servers and blades.
  • Supports running many x86 Linux applications in a Linux on PowerVM.
  • Live AIX and Linux VMs can be moved between servers, eliminating planned downtime.
  • Intelligently flows memory from one VM to another for increased memory utilization.
  • Simplifies the management and improves performance of Fibre Channel SAN environments.
On the con side
  • During high-demand periods, performance can suffer. PowerVM's Linux virtualization implementations have mechanics that allow for very granular resource management and control; during peak periods there is still the potential for performance degradation.
  • With IBM PowerVM, you can virtualize 10 logical partitions (LPARs) to share one CPU or even one NIC; this practice can have a negative impact on performance (too much activity on too little hardware) and availability (consider the consequences of that one CPU failing). The flexibility and configurability of virtualization can lead to poorly designed systems that cause companies to abandon their entire virtualization strategy.
  • Security: In the past, if a server was compromised, the vulnerability could be contained to that one server. With virtualization, every logical partition or virtual environment within the physical server has the potential to be compromised. While a systems administrator has the ability to make sure that the logical partitions within the physical box don't have access to one another, you should not overlook physical security as well.