Re: [RFC] KVM Fault Tolerance: Kemari for KVM

["Fernando Luis Vázquez Cao" <fernando@oss.ntt.co.jp>]

Avi Kivity wrote:
> On 11/09/2009 05:53 AM, Fernando Luis Vázquez Cao wrote:
>>
>> Kemari runs paired virtual machines in an active-passive configuration
>> and achieves whole-system replication by continuously copying the
>> state of the system (dirty pages and the state of the virtual devices)
>> from the active node to the passive node. An interesting implication
>> of this is that during normal operation only the active node is
>> actually executing code.
>>
> 
> Can you characterize the performance impact for various workloads?  I 
> assume you are running continuously in log-dirty mode.  Doesn't this 
> make memory intensive workloads suffer?

Yes, we're running continuously in log-dirty mode.

We still do not have numbers to show for KVM, but
the snippets below from several runs of lmbench
using Xen+Kemari will give you an idea of what you
can expect in terms of overhead. All the tests were
run using a fully virtualized Debian guest with
hardware nested paging enabled.

                      fork exec   sh    P/F  C/S   [us]
------------------------------------------------------
Base                  114  349 1197 1.2845  8.2
Kemari(10GbE) + FC    141  403 1280 1.2835 11.6
Kemari(10GbE) + DRBD  161  415 1388 1.3145 11.6
Kemari(1GbE) + FC     151  410 1335 1.3370 11.5
Kemari(1GbE) + DRBD   162  413 1318 1.3239 11.6
* P/F=page fault, C/S=context switch

The benchmarks above are memory intensive and, as you
can see, the overhead varies widely from 7% to 40%.
We also measured CPU bound operations, but, as expected,
Kemari incurred almost no overhead.

>> The synchronization process can be broken down as follows:
>>
>>   - Event tapping: On KVM all I/O generates a VMEXIT that is
>>     synchronously handled by the Linux kernel monitor i.e. KVM (it is
>>     worth noting that this applies to virtio devices too, because they
>>     use MMIO and PIO just like a regular PCI device).
> 
> Some I/O (virtio-based) is asynchronous, but you still have well-known 
> tap points within qemu.

Yep, and in some cases we have polling from the backend, which I forgot to
mention in the RFC.

>>   - Notification to qemu: Taking a page from live migration's
>>     playbook, the synchronization process is user-space driven, which
>>     means that qemu needs to be woken up at each synchronization
>>     point. That is already the case for qemu-emulated devices, but we
>>     also have in-kernel emulators. To compound the problem, even for
>>     user-space emulated devices accesses to coalesced MMIO areas can
>>     not be detected. As a consequence we need a mechanism to
>>     communicate KVM-handled events to qemu.
> 
> Do you mean the ioapic, pic, and lapic?

Well, I was more worried about the in-kernel backends currently in the
works. To save the state of those devices we could leverage qemu's vmstate
infrastructure and even reuse struct VMStateDescription's pre_save()
callback, but we would like to pass the device state through the kvm_run
area to avoid a ioctl call right after returning to user space.

> Perhaps its best to start with those in userspace (-no-kvm-irqchip).

That's precisely what we were planning to do. Once we get a working
prototype we will take care of existing optimizations such as in-kernel
emulators and add our own.

> Why is access to those chips considered a synchronization point?

The main problem with those is that to get the chip state we
use an ioctl when we could have copied it to qemu's memory
before going back to user space. Not all accesses to those chips
need to be treated as synchronization points.

>>   - Virtual machine synchronization: All the dirty pages since the
>>     last synchronization point and the state of the virtual devices is
>>     sent to the fallback node from the user-space qemu process. For this
>>     the existing savevm infrastructure and KVM's dirty page tracking
>>     capabilities can be reused. Regarding in-kernel devices, with the
>>     likely advent of in-kernel virtio backends we need a generic way
>>     to access their state from user-space, for which, again, the kvm_run
>>     share memory area could be used.
> 
> I wonder if you can pipeline dirty memory synchronization.  That is, 
> write-protect those pages that are dirty, start copying them to the 
> other side, and continue execution, copying memory if the guest faults 
> it again.

Asynchronous transmission of dirty pages would be really helpful to
eliminate the performance hiccups that tend to occur at synchronization
points. What we can do is to copy dirty pages asynchronously until we reach
a synchronization point, where we need to stop the guest and send the
remaining dirty pages and the state of devices to the other side.

However, we can not delay the transmission of a dirty page across a
synchronization point, because if the primary node crashed before the
page reached the fallback node the I/O operation that caused the
synchronization point cannot be replayed reliably.

> How many pages do you copy per synchronization point for reasonably 
> difficult workloads?

That is very workload-dependent, but if you take a look at the examples
below you can get a feeling of how Kemari behaves.

IOzone            Kemari sync interval[ms]  dirtied pages
---------------------------------------------------------
buffered + fsync                       400           3000
O_SYNC                                  10             80

In summary, if the guest executes few I/O operations, the interval
between Kemari synchronizations points will increase and the number of
dirtied pages will grow accordingly.

Thanks,

Fernando