Quick overview of IBM Power Systems
A year ago, as a member of the academy. IBM programs, I attended the IBM AIX 6 Jumpstart for UNIX Professionals course at the Moscow office of IBM. The instructor covered most of the topics so that an experienced Linux or UNIX administrator sat down at AIX and started working, and also described the main points of the IBM Power Systems server line. As it turned out, these are quite interesting Enterprise-level systems that provide complete business systems, including built-in virtualization, monitoring, diagnostics, etc., which are also supported by the operating system.
I do not pretend to be an exhaustive description of architecture, the article is rather an excursion into technology, for general development.
And so, as I said, System P is a line of IBM servers built on the basis of Power processors (RISC architecture). The main purpose is application servers for business systems: database servers, corporate correspondence (for example, Lotus) and miscellaneous middleware (middleware) such as WebSphere.
There are 3 operating systems running on Power servers: native AIX, IBM i, and, of course, Linux.
AIX is one of the oldest UNIX'ov, known for its reliability, stability, flexibility and ease of management (this is mainly achieved through SMIT - console system configurator, with which you can do a lot of things without bothering with command line options). I was shocked at the study of this system by the fact that RPM packages from Linux are installed on it! Under its architecture, of course, but still. It was not difficult to install the native mc on AIX: I downloaded the RPM package for the ppc architecture, downloaded the dependencies, and I installed it all with the well-known Linux team RPM.
IBM i is a separate operating system that is little similar to all known families. I don’t know much about her yet, but friends from the bank who serve AS / 400 with IBM i speak of her as one of the most productive OSes for the DB2 database management system.
Linux for Power is officially supported by only two manufacturers: RedHat, with their RHEL, and Novell, with their SLES. How many did not look for the plate with the description of System P functions which are not supported by Linux, I did not find, but I know that there are such. For example, the fact that ext3, unlike JFS2 from AIX, does not support reducing the size of a partition without unmounting is very unpleasant, therefore, operations with decreasing the LVM of a partition lead to the outage of any services.
The main technological “chips” of System P that distinguish this line from the whole variety: a developed hypervisor, “resources on demand”, flexibility and mobility when building and changing infrastructure, energy efficiency, support for clustering to ensure high availability. Prices, of course, at first glance bite, but in the end, all of the above characteristics and technologies fully cover the initial costs. Now, in order:
Virtualization Tools
This is probably the most interesting part: in Power systems, all OSes work in virtual environments based on a hypervisor. Those. the server can contain several "guests" isolated from each other, working simultaneously and sharing resources. Each logical (virtual server) is called LPAR (Logical Partition Access Resources), in which the operating system is installed. The maximum number of LPARs on a server depends on the number of processors and memory: you can create a maximum of 10 LPARs per processor by allocating a minimum share of 0.1 processor resources and a minimum of 256 MB of RAM for each LPAR. There is also a theoretical limitation: on a 795 model (the last in the line), you can create a maximum of 254 LPARs. Thus, it is possible to optimize the load on the server by dividing the processor pool into virtual servers, i.e.
Each LPAR is also allocated disk space and network cards. This is done by a special server called VIOS (Virtual I / O Server). It consolidates resources: network cards, SAN disk space or local disks, and, according to user settings, gives it all to virtual machines.
But the most pleasant thing is the DLPAR (Dynamic Logical Partitioning) technology: it allows you to dynamically change the amount of resources (processor shares, memory, I / O interfaces) allocated for LPAR dynamically without stopping the system. Those. if I looked at loading partitions and saw that one on average uses 90% of the resources, and the other 30%, I can safely bite off the resources from the second partition without stopping the system and give these resources to the first.
All hypervisor management falls on one of two systems: IVM (Integrated Virtualization Manager) or HMC (Hardware Management Console). IVM comes in a package for VIOS, and provides a convenient web interface for managing the entire server. HMC is already a separate machine with a stripped-down version of linux, which is required to manage multiple servers. It also has a convenient web interface and all servers can be controlled right here. The web interface, of course, can be exchanged for the console, the system provides both options.
Resources on Demand
This is more likely not a piece of hardware itself, but a piece of licensing. It allows you to buy a server with a large number of resources, leaving some inactive, and paying only part of their cost for them, and then, when the server is already in short supply, you can pay IBM for activation and add the necessary resources. You may not need to activate all the resources on an ongoing basis: you can activate them for a while and pay per minute for the resources used. This allows you to add resources at peak times, and not pay for them when the server is relatively idle.
Another rather pleasant point: if, for example, the working processor fails, and there is an unactivated processor, then the working one turns off, and the inactive one turns on, without extra charges.
Mobility
When building the infrastructure, you can use the PowerVM Enterprise Edition. Its main function: Live Partition Mobility. This technology allows servers to forward LPARs to each other without disconnecting them even under load. Of course, it is advisable to use fast channels, SAN arrays, etc., so that it is all fast and customers do not even notice it. At IBM, we were shown how this happens: a loaded LPAR in a matter of seconds “moved” from one building to another, and started working there.
Once with students they conducted an experiment on a virtual box: they took a virtual machine and tried to transfer the virtual machine to another computer via LAN. Then this function was new, it just appeared in the virtual box, and the most interesting thing is that it worked. However, as it turned out later, only if the virtual machine is not loaded. As soon as I load it, the hypervisor performed 10 iterations and stumbled with an error. Of course, a 100 megabit LAN and shared storage in the form of an SMB resource in the class can not be compared with the SAN array and IBM's 10 gigabit optics, but it was worth a try.
Energy efficiency
Power servers have built-in energy and cooling management systems that help save energy and lower costs, provide better performance per watt, and Advanced POWER Virtualization allows you to consolidate servers that are not fully loaded, wasting energy. In addition, EnergyScale technology can provide unique power management capabilities. These features allow users to measure the power consumed by system components and configure policies so that the server environment is efficient in terms of power consumption.
High availability
Power-based servers can be assembled in clusters, providing high availability and cluster multi-processing. This technology was called PowerHA. PowerHA provides automatic fault detection, diagnostics, application recovery, and node reintegration. Together with appropriate software support, PowerHA can provide simultaneous data access for parallel processing applications, thus offering excellent scalability.
Conclusion: servers based on POWER processors have very interesting technological innovations and allow you to build a very flexible and sustainable infrastructure.
From the literature on Power, I advise you to study redbooks. Finding a system to see how it works is quite difficult: the systems are expensive and quite rare, at least with us. Purely human curiosity can be satisfied by watching a video on youtube, or in the documentation on the IBM developerworks site . There is also an excellent Russian site dedicated to Power systems and AIX: aixportal.ru .