First there was data center consolidation, then server, then storage and now . . .
I/O consolidation.
Stealth no more
3Leaf is exiting stealth mode, officially, tomorrow, but they put up their new website over the weekend, so they’re fair game. Incorporated with a $0.5 Million seed round from Storm Ventures in June 2004, raised an A round of $12 Million in April 2005, first betas last May, cleared a $20m B round in September, and are just about ready to start selling product.
Two part harmony
They have a two part product strategy:
- First stage product provides I/O consolidation
- Second stage product provides compute and memory consolidation
They do for I/O what VMware does for servers
Simple cost displacement model: 3leaf claims they’ll give you the same IO for half the price, so the ROI is a matter of weeks. How does that work?
Most servers aren’t working very hard, which means their I/O isn’t either. The idea is to reduce the number of direct server connections to costly enterprise class FC or ethernet switches by consolidating the IO of many servers to a single IO gateway. 3Leaf’s box offers 7 PCI express (and one HT) slots to stick standard HBAs into, while the box connects to the servers through either ethernet or Infiniband. Think backplane extender.
Also think diskless, stateless servers. According to their website, the virtual I/O server:
. . . replaces each compute node’s storage and network I/O with a single, high speed, redundant, and fault tolerant fabric, converting each compute node into a diskless and stateless commodity server with centrally managed bandwidth. Thus, each compute node costs far less to put into service, allows for better control of network bandwidth, and is more flexible, robust, and reliable.
The 3Leaf V8000 associates the I/O state (connectivity, security, quality) with the OS, not the physical server hardware. 3Leaf supports Windows and Linux, requiring a kernal driver for Linux and a Windows equivalent as well as working with VMware virtual servers. The standard configuration is a dual-redundant two-box system.
The next step: Torrenza
I hadn’t heard of Torrenza before. It’s an AMD initiative that puts an empty chip socket on their mobo’s so third parties, like 3 Leaf, can add stuff. From the AMD website:
The Torrenza Innovation Socket enables OEMs who develop their own processors to take full advantage the x86 operating environment. This new approach enables OEMs to consolidate server offerings for multiple processors to potentially a single platform, reducing datacenter disruption and deployment costs for customers.
Leading server OEMs that develop silicon, including Cray, IBM, and Sun, have endorsed Torrenza as an open innovation initiative.
3Leaf is working on a chip for that socket, which will enable them to place those stateless, diskless servers in a virtual server warehouse. As an application needs more stuff, the 3Leaf mojo will dole it out in real time. Sounds cool to me.
The StorageMojo take
3Leaf’s CEO Bob Quinn is clear that this is for enterprises, not scale-out internet data centers. Also for data center service providers, for example Savvis, a 3Leaf beta customer. Which tugs on some interesting threads:
- Enterprise computing growth is slower than Moore’s law – so consolidation is the new normal for the enterprise.
- 3 Leaf is yet another shot across the bow of costly FC and ethernet infrastructure. As Bob later wrote to me: “3Leaf will increase SAN utilization through economic connectivity for commodity X86 servers.” True enough. Which also says that fewer ports will be sold in 3Leaf equipped data centers.
- Is somebody going to wake up and say, “Why the hell don’t we build an OS that can handle lots of concurrent applications?” It just feels like virtualization is fixing problems created by hinky software and 30 year old paradigms. Anyone for a clean-sheet re-think of application environments?
I’m keeping an open mind about Torrenza – Intel seems to have a parallel idea as well – and how it might play out. On its face, a great idea. Yet it could also be seen as turning a commodity into a non-commodity. Which is obviously the intention, and could be its undoing.
If smart people develop cool stuff and market it well, this could offer value-add that people flock to. On the other hand, if everybody treats it as an opportunity to shake more money out the enterprise they might be disappointed.
Sure, the decline in enterprise server cost frees up money for other toys. But CFOs aren’t ignorant and it won’t be long before they are trying to squeeze savings from the shrinking IT infrastructure. 3Leaf is well-positioned to catch that wave.
Update: Bob went through the article and offered some additions and clarifications that improve the information value of the post. I’ve added some but not all of them. Anything that is still unclear or stupid is my fault. I did check their website for whether 3Leaf has a space or not and got that wrong all by myself.
Comments welcome, as always. Seems like there are several topics here worthy of thoughtful dialogue.
Robin,
Another ‘stealth’ website… not much detail.
You say “yet another shot across the bow of costly FC and Ethernet infrastructure” … I hope it is PCI Express and not Infiniband.
I hope they have an easy, low cost way to connect to storage…. this may take another round of VC funds.
It seems that the cost of the 10Gbit Ethernet proving to be as painful as the cost of FC ?.
This sounds the same as Topspin VFrame.
So it’s cheaper to buy the 3Leaf gateway and an IB HCA for each server than to buy a FC HBA for each server? Or is the idea to spend money on IB switch ports and save money on Ethernet/FC switch ports? Maybe I could imagine that, since enterprisey switches are so expensive.
BTW, I think the OS that can handle lots of concurrent applications is supposed to be Solaris.
My impression is that Richard and Wes are having at least some of the same difficulties seeing any major value-add here that I am. If a server isn’t working its storage all that hard, surely using 1 to 4 GigE ports into the fabric will be far less expensive (and possibly more performant) than introducing an additional smart box into the path, and if its needs are greater than that, where’s the benefit – is there some real reduction in *management* overhead here?
I also agree with Wes about Solaris (and perhaps AIX and even crusty old HP-UX as well; *certainly* VMS, though at this point, perhaps no one cares any more): there’s no lack of OSs that handle lots of processes (whether homogeneous or heterogeneous) just dandy – Windows just doesn’t happen to be one of them, and that’s the problem, since porting applications to a more competent OS appears to present more obstacles than simply virtualizing Windows systems such that each OS instance need support only a single application.
The other related point is that the old Unix paradigm of one process per client connection is pretty much passe: multi-threaded applications are common now virtually everywhere they are appropriate, which would eliminate a fair amount of the value of virtualizing servers if Windows itself were as reliably stable running such an app as some of the more mature OSs are (and thus didn’t need virtualizing to minimize the number of threads lost on a failure).
Some people believe that Windows really can come somewhere near matching the reliability of some of the other OSs I mentioned if properly configured. If so, then ‘proper configuration’ appears to be something that few Windows admins are able to achieve, creating much of the thriving virtualization market which we see today (the balance perhaps coming from economies in management that can accrue from managing only a single application per server rather than suitably configuring a single OS instance to protect multiple applications from each other).
– bill
To elaborate on my point about Solaris, Zones are supposed to allow the “one application per OS” model to be used with no performance overhead or extra license fees. I think handling concurrent applications is not really about performance or reliability or stuff like that — it’s about managing configurations (such as “app A requires Java 1.4.2.7.5r2 but app B requires Java 1.5.6.8”).
3Leaf’s approach seems to be to efficiently run one app per (cheap) server but soon even a low-end server will be so powerful that you’ll need some kind of virtualization inside the box. At that point it becomes a fight between the OS and hypervisor vendors.
Modern applications are mostly developed under a virtual machine model (e.g. Java or .Net).
Recently vendors such as Weblogic and IBM have realised that running a virtual application machine on an (bloated) OS which is in turn running on a virtual hardware machine is just silly. I expect that we will soon see apps that run a lot closer to the base virtual machine (though they won’t know or care), cutting out lots of fat and overhead (you can run a slim linux VM with Java in less than 100MB).
Thus most of the heavy lifting will be in the VM hypervisor and its drivers, which will also manage the sharing with the applications.
I can also see DB VMs that cut most of the OS out (maybe part of why Oracle is so keen on Red Hat?) and again take on most of the storage smarts.
The hypervisor could then provide two interfaces, a file based one (like ZFS or Google) and an object based one like you have mentioned before.