What is it?
One of the cool new features that takes scalability in Windows Server 2012 R2 Hyper-V to a new level is virtual Receive Side Scaling (vRSS). While since In Windows Server 2012, Receive Side Scaling (RSS) over SR-IOV is supported it’s best suited for some specialized environments that require the best possible speeds at the lowest possible latencies. While SR-IOV is great for performance it’s not as flexible as for example you can’t team them so if you need redundancy you’ll need to do guest NIC Teaming.
vRSS is supported on the VM network path (vNIC, vSwitch, pNIC) and allows VMs to scale better under heavier network loads. The lack of RSS support in the guest means that there is only one logical CPU (core 0) that has to deal with all the network interrupts. vRSS avoid this bottleneck by spreading network traffic among multiple VM processors. Which is great news for data copy heavy environments.
What do you need?
Nothing special, it works with any NICs that supports VMQ and that’s about all 10Gbps NICs you can buy or posses. So no investment is needed. It’s basically the DVMQ capability on the host NIC that has VMQ capabilities that allows for vRSS to be exposed inside of the VM over the vSwitch. To take advantage of vRSS, VMs must be configured to use multiple cores, and they must support RSS => turn it on in the vNIC configuration in the guest OS and don’t try to use a home PC 1Gbps card
vRSS is enabled automatically when the VM uses RSS on the VM network path. The other good news is that this works over NIC Teaming. So you don’t have to do in guest NIC Teaming.
What does it look like?
Now without SR-IOV it was a serious challenge to push that 10Gbps vNIC to it’s limit due to all the interrupt handling being dealt with by a single CPU core. Here’s what a VMs processor looks like under a sustained network load without vRSS. Not to shabby, but we want more
As you can see the incoming network traffic has the be dealt with by good old vCore 0. While DVMQ allows for multiple processors on the host dealing with the interrupts for the VMs it still means that you have a single core per VM. That one core is possibly a limiting factor (if you can get the network throughput and storage IO, that is). vRSS deals with this limitation. Look at the throughput we got copying lot of data to the VM below leveraging vRSS. Yeah that’s 8.5Gbps inside of a VM. Sweet . I’m sure I can get to 10Gbps …
At TechEd 2103 Europe I met some passionate people who’ve been working hard at deploying Storage Spaces on a large scale in their companies (many hundreds of TB). They have replaced their expensive SANs and getting better performance, better data protection leveraging ReFS at way cheaper pricing (licensing is a killer). Their life has also become easier due to the KISS principle and the fact that complex solutions have complex issues.
Storage Spaces are creating a buzz. Lot of interest, interactive discussions & paper napkin designs of concepts were being drawn up. It seems like more and more people are getting the message and don’t consider my talks over the past 18 months on this as weird any more. So you need some “out of the box” thinking to get “Cluster in a Box” as a building block for scale out storage solutions. I’ve already done away with SAN built on proprietary hardware, it’s all commodity gear. But t doesn’t stop there. Once you realize that this hardware gives you all you need the logical step is the software and that’s where Windows and Storage Spaces comes in. Yes sure, this is blasphemy in the storage world, so what? The world used to be flat once
There is no singe universal solution. Just good to great ones. Don’t fall for the FUD big storage vendors fling around and don’t be man handled into a solution. The reality is that when you have what works for you, that’s when you’ve got it right. So look into Storage Spaces and all related technologies. Once you wrap your head around the concepts, you start to understand the why an how of it all a lot better.
Whatever the nature and the size of your environments and needs, there are options or even combinations of options out there that can help you. You can buy completely configured building blocks (cluster in a box) or even complete Fast Track solutions based on this concept with Hyper-V and the System Center stack. You can also build the solution your self. Combine this with total VM mobility within the data center in combination with continuous availability. Outside of the data center Hyper–V Replica and application based shared nothing high availability help failover fast with out noticeable or minimal service interruption. All this without the “Kings Ransom” prices you have to pay for proprietary solutions. If you’re not interested yet I bet your business and especially your CFO is! All you got to do is manage risk instead of letting fear rule you.
How crazy am I with in this view? Some people actually asked me that combined with the question if anyone is really considering using this is real life. Well I met a lot of like minded people who are putting into practice here at TechEd. And knowing my Hyper-V, Clustering, Networking and storage very well, discussing these subjects and Storage Spaces with other attendees I’ve had a job/consulting offer just about every day. Not too bad as far as real life interest in this goes (and I remember being ridiculed in 2008 for being an early adaptor of Hyper-V ).
The storage world is in flux sending waves through the way we deliver our services. Surf the waves and enjoy the ride .