As we all know there are a number of ways of scaling capacity in a vSAN environment, you can add disks to existing hosts and scale the storage independently of compute, or you can add nodes to the cluster and scale both the storage and compute together, but what if you are in a situation where you do not have any free disk slots available, and / or you are unable to add more nodes to the existing cluster? Well vSAN 7.0U1 comes with a new feature called vSAN HCI Mesh, so what does this mean and how does it work?
Let’s take the scenario below, we have two vSAN clusters in the same vCenter, Cluster A is nearing capacity from a storage perspective, but the compute is relatively under utilised, there are no available disk slots to expand out the storage. Cluster B on the other hand has a lot of free storage capacity but is more utilised on the compute side of things:
Now the vSAN HCI Mesh will allow you to consume storage on a remote vSAN cluster providing it exists within the same vCenter inventory, there are no special hardware / software requirements (apart from 7.0U1) and the traffic will leverage the existing vSAN network traffic configuration.
This cool feature adds an elastic capability to vSAN Clusters, especially if you need to have some additional temporary capacity for application refactoring or service upgrade where you want to deploy the new services but keep the old one operational until the transition is made.
VMware has not left the monitoring capabilities of such use out either, in the UI you can monitor the usage of “Remote VM” from a capacity perspective as well as within the performance service
So this clearly allows dissagregation of storage and compute in a vSAN environment and offers that flexibility and elasticity of storage consumption are there any limitations?
A vSAN cluster can only mount up to 5 remote vSAN Datastores
The vSAN Cluster must be able to access the other vSAN cluster(s) via the vSAN Network
vSphere and vCenter must be running 7.0U1 or later
Enterprise and Enterprise Plus editions of vSAN
Enough hosts / configuration to support storage policy, for example if your remote cluster has only four hosts, you cannot use a policy which requires RAID6
So this is a pretty cool feature and sort of elliminates the need for Storage Only vSAN nodes which was discussed in the past at many VMworlds
As we all know storage media has evolved very quickly over the past few years, the decline of the spinning disk and the move to flash based storage devices, but also the shift from SAS/SATA protocol based drives to NVMe protocol based drives in order to address the performance limitations of older protocols that were designed for spinning disks and not for SSDs.
A question I get asked regularly is what type of SSD is best for the vSAN Cache tier, there are vSAN ready nodes out there that contain NAND based SSDs for both the cache tier and for the capacity tier, but then there are other technologies like Intel Optane™ SSDs being used for the cache tier, so let’s talk about the two, for the purpose of this comparison I am going to use the most common 3D NAND based NVMe in vSAN Ready node configurations, the 1.6TB Intel P4610 NVMe drive, and the 375GB P4800X Intel Optane™ SSD, both of these SSDs are NVMe based devices.
As you can see from the above table there are some major differences between the two different SSD’s notably the Random Write performance which is critical in the cache tier in a vSAN environment as all the incoming writes are random in nature and are absorbed by the cache tier, the NAND based SSD does not have as much capability around the random writes versus the Optane™ SSD, but the biggest impact to a vSAN Cache tier is the Drive Writes Per Day (DWPD), if you look at the specifications in detail, the P4610 can handle around 3.5 DWPD which equates to around 5.6TB of data written daily, whereas the 375GB Optane™ SSD can handle up to 60 DWPD which equates to 15TB of data written daily, remember that the Optane™ SSD is also less than a quarter of the capacity of the P4610, so in a vSAN environment cache tier, the Optane™ SSD wins hands down from an endurance perspective as well as the abaility to handle the random writes a lot quicker, so why such a difference?
Well if you look at NAND based SSDs, firstly there is usally an element of DRAM that acts as a buffer to the NAND media which is usually around 1GB of DRAM for every TB of media, so any incoming writes hit the DRAM buffer first, this can be a positive boost in short, low block size write bursts, but cannot be sustained over a longer period of time, in an Optane™ SSD there is no such DRAM buffer so the data is being written to directly to the media. The VxRAIL team at Dell EMC have done some extensive testing around this and clearly demonstrated that a NAND based SSD cannot sustain the same level of write performance in a continuous fashion whereas the Optane™ SSD maintains the same level of write performance consistently, below is the results of their performance testing:
The way NAND based SSDs and Optane™ SSDs perfrom write operations is fundamentally different, in everybody’s NAND, media has to be read and written in pages, but everything has to be erased in blocks. Page updates are typically written to a new unused block, as new data is written, old pages become stale, and on an SSD these stale pages can build up fairly quickly which means at some point there a significant chunks of blocks that are obsolete, this then has to be garbage collected. This will then clear the block and allow that block to receive data, and the process starts all over again.
Optane™ SSDs are transistor-less which essentially means that each cell state can be changed from a 0 or 1 independently of other cells on the device. This means that Optane™ SSDs are completely bit addressable as opposed to having to write in pages, there is also no garbage collection required, and this obviously has a positive impact on performance as well as endurance which is why Optane™ SSDs have very high endurance capabilities.
So what does all this mean from an application perspective? Well the VxRAIL Guys at Dell EMC also did some performance testing using Hammer DB and shown some significant performance gains when using Intel Optane™ SSDs versus traditional NAND as Cache as much as a 61% gain in performance in a complex OLTP workload
As we all know latency is critical in any type of workloads, what I have seen in performance testing is that Intel Optane™ SSDs consistently provide lower latency as well as a much more tightly controlled standard deviation on latency versus the P4610, even though in some smaller block size tests the performance of both devices was similar, in larger block size tests the Optane™ SSD again delivered lower latency and tightly controlled standard deviation in latency but also provided a much higher performance in comparison to the P4610. You also have to remember that the P4610 device was only using 37% Span due to vSAN Currently having a limit of 600GB write buffer per disk group, whereas the Optane™ SSD was using 100% Span, so the P4610 had a bit of an unfair advantage here.
Conclusion What is clear from a vSAN perspective, endurance plays a critical role in the vSAN cache tier, in the very early days of vSAN there was no other choice but SAS or SATA based NAND devices with a ranging DWPD of between 10 and 25 based on an 800GB Drive, but as the technology evolution pushes the boundaries of performance and endurance, technology like Intel Optane™ SSDs clearly have an edge offering up to 60 DWPD on a smaller capacity of 375GB.
Smaller cache device…are you serious?
In the testing I have done on full NVMe systems where Intel Optane™ SSDs are being used in the vSAN Cache Tier, and standard more read-intensive NVMe drives like the Intel P4510 are being used in the capacity tier, a 375GB Optane™ SSD is more than sufficient, in most workloads a 750GB Optane™ SSD did not improve performance, even with 375GB I was only able to saturate the write buffer by 60% (based on vSAN 6.7 Update 3).
So whilst NAND based devices are fully supported as a vSAN cache device, they may not be the right choice when it comes down to consistent performance and endurance required for a modern infrastructure.
As the core density increases on a CPU it opens up the opportunity to consolidate the number of nodes required in any given cluster, but in a vSAN cluster, node consolidation has a negative effect on available IOPS, if you think about how each node provides a specific amount of IOPS, lowering the number of hosts in the cluster removes the IOPS capability of the nodes you are consolidating by, take the following for example:
Number of VMs : 200 vCPU Per VM : 4 Virtual Memory per VM : 32GB Storage per VM : 600GB vCPU to Core Ratio : 4 to 1
Now for the purpose of this sizing excersize I am going to use the vSAN Sizing tool and apply some cluster settings as per below:
So in the above scenario, the number of cores per CPU is 18, and I want to ensure that this is a two disk group configuration, if we then input the workload details as per below:
You will see when we click on recommendation that it shows a required node count of 8 (not taking into account any N+1 capability as we left that as 0 for the purpose of this sizing)
And we can see the disk config below:
However, if we increase the number of CPU Cores to 20 by clicking on the “+” in the sizing output we can see that it changes the number of nodes
And again if we increase the number of cores again to 22 we get a further reduction in the number of nodes to 6
The sizing tool will dynamically increase or decrease the number of disks required per host as well as the RAM per node that is required as you can see here:
But one thing we have not factored in here is the decrease in IOPS Capability that reducing by two nodes , if say for example each node was capable of 80K IOPS, reducing the node count by two means you have just lost 160K IOPS Capability, so what can we do to mitigate that?
Well instead of using SAS/SATA SSDs in your vSAN design, you could opt to use Intel Optane for Cache, and NAND based NVMe drives for capacity. For write operations, Intel Optane greatly improves on write performance as I have written about before, but also read performance is greatly accelerated because the capacity devices are NVMe, so therefore reducing your node count by two in this case and utilising this kind of technology means you still get similar levels of performance, the best part is, the overall solution will cost you less too, so your TCO comes down which is good for your finance department right?
One question I get asked frequently is what size Optane device is sufficient?
Well in all of my testing, I very rarely saturated the write buffer even with 375GB Optane drives as cache devices, the reason for this is because vSAN starts to perform de-staging from the cache tier to the capacity tier when the write buffer becomes around 30% Full, and because the capacity tier it NVMe based, the de-staging happens a lot quicker, especially since vSAN 6.7 U3 where the de-stage limits have been removed.
So when would a 750GB Optane be useful?
High write intensive workloads such as Video Surveilance and Databases, or when your capacity disks are much slower, Optane can still be used in vSAN Configurations where the Capacity Tier is SAS/SATA which of course are not as fast as most NVMe devices so the write buffer can get more full.
So just to re-cap, you can save money on your vSAN deployments by consolidating hosts with higher core count CPUs as well as leveraging newer technology such as Intel Optane in the Cache Tier and NVMe in the capacity tier thus saving money whilst maintaining same level of performance or better, what’s not to like?