CPU

In my earlier post, I discussed some of the main concepts, challenges and techniques that are essential for virtualizing a CPU or instruction set.  To a large degree, these concepts and techniques are not new – they are really just variations on a theme that has been around for a long time in computer science.  Virtualization is fundamentally about preserving the appearance of isolating resources, while actually sharing the resources harmoniously and efficiently.  Perhaps the most obvious example was virtual memory – I can still remember being excited about virtual memory when I upgraded to System 7 on an Apple Quadra!  Tricks with virtual memory would let you run applications without buying new RAM (although the loading performance was often horrible as a result…at least it ran).

 

One of the most difficult and complicated areas of a modern PC is the I/O architecture.  The I/O architecture governs how the processor and memory talk to the devices that make a computer interesting – keyboards, mice, network cards, GPUs and hard drives.  It’s essential to remember that I/O is really a first class priority of a computer, because as a CPU architect, I/O moves at a glacial speed. 

 

Modern CPUs operate at around 3GHz, so a single cycle is only 0.33ns.  In comparison, reading data from a disk takes around 5ms – or 15 million cycles!  The engineering maxim of ‘make the common case fast, and the uncommon case correct’ is sometimes erroneously simplified into ‘ignore the uncommon case’ – and when I/O only happens every 15 million cycles, it’s pretty uncommon. Read More »

Virtualization is one of the oldest and most fundamental concepts in computer science and IT; if you look carefully, it is almost everywhere – desktops, notebooks, servers, SANs, cell phones, even GPUs!

Virtualization entails replacing a dedicated computing or storage resource (e.g. CPUs) in a system with an abstraction of that resource (e.g. a virtual CPU) that appears to be dedicated, but is actually shared between several uses and managed to achieve higher efficiency. The appearance of being dedicated (i.e. isolation) is key because dedicated resources are very easy and convenient to use; there’s no concern that someone else might be hogging that resource.

This sounds deceptively simple, but the reality is quite complex. Many times, different elements of a system will conflict with one another – those conflicts must be seamlessly arbitrated. Even if this is done correctly, sometimes the performance overhead is so bad that the result is almost unworkable.

Virtual memory is a particularly relevant example. To virtualize memory, there are two key issues – resources and how to access them. Typically virtual memory is broken down into two sections – kernel space, which is a private region of memory for ring 0 (i.e. the most privileged) software – that is the OS kernel, IO devices, drivers and a few other odds and ends. Kernel space cannot be paged out to disk, so there are no page faults in the kernel. This restriction is one reason why most kernels are small. User space is made available to ring 3 software (aka applications, the least privileged) and can be paged out. The OS provides and a separate user space for each application, which is how they are protected from each other – so that bad data in one application will not cause problems elsewhere, and that applications cannot maliciously or accidentally access each other’s data. Read More »

First dual-core in 2005, then quad-core in 2007: the multi-core snowball is rolling. The desktop market is still trying to find out how to wield all this power; meanwhile, the server market is eagerly awaiting the octal-cores in 2009. The difference is that the server market has a real killer application, hungry for all that CPU power: virtualization.

While a lot has been written about the opportunities that virtualization brings (consolidation, hosting legacy applications, resource balancing, faster provisioning…), most publications about virtualization are rather vague about the “nuts and bolts”.

Performance? Isn’t that a non-issue? Modern virtualization solutions surely do not lose more than a few percent in performance, right? The answer is quite a bit different from what some of the sponsored white papers want you to believe. Read More »