Orginal Article By: Ken Milberg (mailto:kmilberg@gmail.com), Writer/site expert, Future Tech Most IBM® AIX® administrators understand the virtualization features available to them on their System p® platform through PowerVM™, which is also available on the System p for Linux®. But what about the other UNIX® hardware platforms? What do they have to offer and how do some of their features compare to PowerVM?
HP virtualization This section discusses HP's Virtual Server Environment (VSE) as well as vPars, nPartitions, and IVMs.
HP's VSE is the front end for HP's overall virtualization strategies. VSE itself contains several elements, including a workload management tool and advanced manageability software. The workload management feature allows you to draw from spare capacity, which is available on a pay-per-use basis. This is similar in many respects to IBM Capacity on Demand offerings. Further, HP Global Workload Manager (gWLM) provides intelligent policy engines that allow for automatically adjusting the workloads to increase server utilization. It also comes with a product called HP Capacity Advisor that helps you simulate various workload sceneries, which is similar in many ways to the IBM System Workload Estimator (WLE), which ships with the System Planning Tool. The VSE also allows you to partition in several ways, with hard and soft partitions, as well as HP Virtual machines, partitions, and secure resource partitions. HP describes their virtualization/partitioning solutions as their partitioning continuum. Available partitioning includes:
- nPartitions offers true electrical isolation, as well as cell granularity. nPartitions are based on hard partitions, which were first introduced by HP in 2000 and offer greater fault isolation than soft partitions. nPartitions allow you to service one partition while others are online, which is similar to IBM logical partitioning, though systems require a reboot when moving cells from one partition to another. It's important to note that while nPartitions support HP-UX, Windows®, VMS, and Linux, they only do so on the Itanium processor, not on the HP9000 PA Risc architecture. Partition scalability also depends on the operating system running in the nPartition. Another downside is that entry-level servers do not support this technology, only HP9000 and Integrity High End and Midrange servers. They also do not support moving resources to and from other partitions without a reboot.
- vPars are separate operating system instances on the same nPartition or server. This offering allows you to dynamically move either CPU or memory resources between partitions as the workload requirements change. They also allow you the ability to run multiple copies of HP-UX on the same hardware. Using vPars, you can move CPUs to other running partitions, similar to PowerVM and the System p. What you can't do with vPars is share resources, because there is no virtualized layer in which to manage the interface between the hardware and the operating systems. This is one reason why performance overhead is limited, a feature that HP will market without discussing its clear limitations. The scalability is also restricted to the nPartition that the vPar is created on; the max being an 8-cell limitation. There is also limited workload support; resources cannot be added or removed. Finally, vPars don't allow you to share resources between partitions, nor can you dynamically allocate processing resources between partitions.
Integrity Virtual Machines (IVMs) are separate guest instances on the same nPartition with different operating system versions and users in a fully isolated environment. First introduced in 2005, they allow for a partition to have its own full copy of the operating system. Within this copy, the virtual machines share the resources. This is similar in many ways to IBM PowerVM, as there is granularity for CPUs and I/O device sharing. The granularity actually beats PowerVM as one can have up to 1/20 of a micropartitions; the System p allows for only 1/10 of a CPU. The downside here is scalability. With HP's virtual machines, there is a 4 CPU limitation and RAM limitation of 64GB. Reboots are also required to add processors or memory. There is no support for features such as uncapped partitions or shared processor pools. Finally, it's important to note that HP PA RISC servers are not supported; only Integrity servers are supported. Virtual storage adapters also cannot be moved, unless the virtual machines are shut down. You also cannot dedicate processing resources to single partitions. Resource Partitions are created from the HP Process resource manager and allow resources for specific applications within a single operating system. This is a resource management tool that allows you to manage CPU, memory, and disk bandwidth. It allows minimum allocations of CPUs and even allows you to cap a CPU by group. In many ways this is similar to a Solaris containers or AIX WPARs in that it allows you to have several applications reside in one copy of HP-UX. This feature has been available since HP-UX 9.0.
Sun virtualization Now that we've looked at HP virtualization, let's take a close look at Sun's xVM, along with containers, hardware partitioning, and logical domains.
Sun has made some bold changes during the past year, the most important being the release of xVM. Sun's xVM is actually a mix of four separate technologies, the first of which was introduced in February of 2008: xVM Ops Center. xVM Ops Center's most important function is that it provides a single console for the management of all devices in a virtualized environment. It further allows for the discovery and management of all physical and virtual assets. The other three technologies include xVM server, VirtualBox, and VDI software. The xVM server is a hypervisor-based solution, which is based on Xen, running under Solaris on x86 computers. On SPARC it is still based on logical domains. Containers and LDOMs now are part of the umbrella named xVM. Sun is also claiming features such as predictive self-healing, which has long been available on the System p. VirtualBox is desktop virtualization software geared toward developers, allowing for many different types of operating systems to run on top of an existing desktop operating system. It supports Windows, Linux, Mac, and Solaris hosts. Sun did not develop this product, but acquired open source desktop virtuaization vendor Innotel, which develops the product.
Sun also offers hardware partitioning, which allows their high-end servers to be divided into four-process partitions. These are referred to as Sun DSD's. In many ways this technology is similar to IBM logical partitioning, which was introduced in 2001, with no real virtualization capabilities. It is also similar to HP's hardware partitioning, but only high-end and midrange servers support this technology. You cannot share resources between partitions, nor can you dynamically allocate processing resources between partitions. You also cannot share any I/O. It's the LDOMs that actually allow virtualization. Introduced in 2007 on their SunFire line of servers, LDOMs enable customers to run multiple operating systems simultaneously. While LDOMs solved a huge deficiency in Sun's virtualization strategy, it has many inherent flaws:
- Scalability -- Only eight CPUs and 64GB RAM on one machine
- Server-line -- Only low-end Sparc servers are supported
- Limited micro-partitioning -- Four partitions on T1, Eight on T2
- No Dynamic allocation between partitions
For years, Sun's answer to everything was containers or zones. Containers and zones give you the ability to run multiple virtual operating system instances inside only one kernel. They are used to provide an isolated and secure environment for running applications, which are created from a single instance of Solaris.
Simply put, they had it and IBM did not. Sun can no longer make this claim. IBM now offers AIX workload partitions (WPARs), which is their answer to containers. IBM WPARs have all the features of containers plus additional innovations:
- Application WPARS. This is a workload partition that allows a single process or application to run inside of it, like a wrapper. Unlike a standard WPAR, it is temporary and stops when the application ends.
- Live Application Mobility. This feature allows you to move running WPARs to other partitions without any user disruption. With Solaris, you need to shut off the zone first. The feature also allows you to perform multi-system workload balancing, which can be used to conserve data center costs.
While Sun appears to be moving in the right direction with xVM, it still needs to standardize its offerings more. There is also still too much confusion around their offerings and virtualization roadmap.
IBM virtualization IBM has a 40-plus year history of virtualization. No other vendor can come close to making this claim. The fact is that virtually (pardon the pun) everything they have implemented on the midrange, they have already done on the mainframe. They offer one virtualization strategy, PowerVM, unlike the myriad of solutions available from Sun or HP. The technology itself uses a hypervisor based solution (which IBM has finally implemented though Xen, but only on their x86 platform), which sits between the operating system and the hardware. PowerVM is a combination of hardware and software, which accounts for IBM Systems p virtualiation strategy.
The technology that makes up PowerVM includes:
- Micropartition and Shared processor pools. Micropartioning allows you to slice up pieces of your CPUs into virtual partitions. At the same time, it allows for the sharing of CPU, RAM, and I/O. You can carve your partition with up to 1/10 of a CPU. Using a feature called uncapped partitions, you can even exceed the amount of hardware that you configure on your partition, your entitled capacity (EC), which is an important feature in environments where activity fluctuates. This workload management is all done automatically, without requiring the usage of special workload management tools. IBM has since retired their workload management tool, Partition Load Manager (PLM), recognizing that it was the automation inhertent in its shared processor pool strategy, which really captivated the audience.
- VIO servers. These are special partitions that allow you to service resources to VIO clients. The servers own the actual resources, which are network adapters or disk I/O. These partitions save money and provide flexibility by allowing partitions to shared I/O resources. Shared Ethernet and virtual SCSI are the solutions that allow for sharing network and disk I/O.
During the past year, IBM has changed their virtualization nomenclature from Advanced Power Virtualization to PowerVM. In the process it has added several significant features to their virtualization product line:
- Live Partition Mobility. This feature, introduced with their Power6 architecture, allows you to move running AIX or Linux partitions from one physical server to another.
- Lx86. This recent innovation allows you to run x86 Linux applications that were not specifically ported to the Power Architecture directly on a Linux partition without a recompile.
- Shared Dedicated Capacity. This feature allows you to use spare CPU cycles from dedicated processor partitions towards the shared processer pool.
- Integrated Virtualization Manager (IVM) -- This is a browser-based tool that provides the option of not having to have a dedicated hardware management console (HMC) from which to manage virtualization resources.
Summary This article discussed the offerings of the top UNIX vendors and compared them to IBM. It showed how most of the features available from HP and SUN are already implemented by IBM on a much grander scale. It also showed how IBM solutions are more scalable and mature. While IBM continues to innovate and improve on their offerings, it seems that the other vendors continue to just play catch up. While HP has more offerings and a more mature set of a virtualization feature sets than Sun, it still lacks the scalability and flexibility of IBM PowerVM. While Sun has a new name and has made some bold changes in the past year, they still have a long way to go.
via [ IBM ]
|
Posts
