Software house Atlantis has updated its ILIO product, which juggles the storage of virtual desktops in data centres, so that virtualised machines run entirely in a server's memory - sans a SAN or local drives. Atlantis ILIO Persistent VDI 4.0 lets Citrix XenDesktop and VMware View customers run persistent virtual desktops in- …
Sounds good but...
What happens when you need to reboot a server, the mains power dies and you're stuck with a few minutes of UPS (assuming it works) or a system crashes? How often is everything actually written to some form of persistent storage?
I can see a power failure taking a business out for days or weeks, rather than just a few hours or one day if you are unlucky enough to have an unexpected power failure. And you will experience one, one day.
If *everything* is written regularly it might be quite nice, but then you wouldn't get the drastic reduction in storage traffic or storage utilisation they are claiming.
Re: Sounds good but...
The clever bit about the Atlantis software (when I looked at it in anger a year or so ago after seeing it at VMworld anyway) was that it did two things to take the load off of backend disk - firstly, it did full deduped RAM caching of disk blocks in the VMhost RAM, (for both read and write) - the dedupe making it very efficient as most of the OS image hot blocks are common.
Secondly, it did dedupe, serialisation and compression of any writes back to disk, using an "inline" virtual NFS store - basically it shows your VMhost an NFS share (which live sin memory), where you provision your VMDK's as normal, but that NFS actually lives on an Atlantis VM on each host, which is then synced back to the underlying disk.
The idea is you should use the Atlantis provisioned storage for your VDI boot images and working drives, and stick your "proper" data on more traditional storage (file server or NAS or whatever), giving your very fast desktops with reliable file storage.
There's a good writeup of the new release (with comments input from the Atlantis guys) on Brianmadden.com if you want more detail. They've solved the persistant VDI image restriction, and it now supports multiple hypervisors etc. Nice that its licensed per desktop, and I reckon a per host version for server VM accelerationis forthcoming soon...
Sounds good but...
Hi Skoorb – The Atlantis ILIO Fast Replication technology, new in Atlantis ILIO Persistent VDI 4.0, constantly synchronizes the state of the virtual desktops to persistent, shared storage in a consistent manner. While committing the state of the desktop to storage, Atlantis ILIO applies two tiers of optimization (context-aware IO processing, inline deduplication, compression, coalescing, etc.), once on the host where the virtual desktops are running in RAM and once centrally across multiple virtual desktop hosts. This reduces the network bandwidth and storage performance and capacity requirements by up to 95%.
The virtual desktops themselves do not interact with the storage directly, but rather run straight out of server RAM and provide a user experience that is better than a PC or laptop, even if they are equipped with an SSD.
So, if there is a server failure due to a power outage or system failure, the hypervisor’s High Availability functionality will automatically restart the virtual desktops and the Atlantis ILIO virtual machine on a standby server. Atlantis ILIO enables the standby host to quickly rebuild the local server RAM datastore from shared SAN/NAS storage.
Hi Phil – Thanks for the feedback. Stay tuned for more details on Atlantis ILIO for Servers!
RAM bandwidth. It was already a thing with virtualization at the levels we can get with today's servers. With this...? What is the memory controller made out of? Unicorns?
Within Atlantis Computing, we like to rever to it as pixie dust, as it makes VDI fly! :)
For more information on the product and how it actually works, please visit Andrew Wood's blog (http://www.virtualizationpractice.com/atlantis-ilio-persistent-vdi-4-0-vdi-game-changer-20573/) or listen to our Director of Marketing Seth Knox' podcast with Doug Brown (http://www.dabcc.com/media.aspx?id=2282)
+1 to marketing for the witty response. I'll check out the resources mentioned in the hopes they answer my question. I'm hoping we're not simply being asked to substitute one bottleneck for another...
Trevor, feel free to send me email at my first name AT atlantiscomputing DOT com if you have follow-up questions.
- Analysis iPhone 6: The final straw for Android makers eaten alive by the data parasite?
- First Crack Man buys iPHONE 6 and DROPS IT to SMASH on PURPOSE
- First Fondle Reg journo battles Sydney iPHONE queue, FONDLES BIG 'UN
- TOR users become FBI's No.1 hacking target after legal power grab
- Vid Reg bloke zips through an iPHONE 6 queue from ZERO to 60 SECONDS