Room: Mandalay Bay L, Level 2
In this first session, Duncan and I will delve deep into the simplicity and benefits of Storage Policy Based Management (SPBM for short). SPBM is the glue that allows vSAN, Virtual Volumes, IO Filters, and even Project Hatchway (persistent volumes for containers) to consume storage with specific data services on a per VM or even VMDK basis. We have lots of demos and will explain how this feature is key to our Software Defined Storage vision at VMware. If you’re looking to understand some of the benefits of vSAN or VVols over traditional storage, this might be the session for you.
This next session is based on quite a bit of research that Paudie and I carried out earlier this year. A common question from our customers continues to be around getting vSAN optimized for performance, especially during a proof of concept/benchmark. We will cover the basics of prepping vSAN for a benchmark, and then take you though a sample setup and test run using HCIBench, a benchmarking tool designed for scale-out HCI solutions, and based on the Oracle vdBench tool. We will then show you some of the different results we achieved when adding extra disk groups, stripes, etc. More importantly, this session will call out any gotcha’s you might encounter, based on what we have seen in the field. Basically, it will prepare you for a successful performance benchmark on vSAN.
This final session is an interesting one. We already have many customers running cloud native, next-gen apps on top of vSAN. We will talk about some of these initially, but mainly we will focus on some of the challenges and considerations around doing this today. We will then focus on what is coming down the line to make this somewhat easier and more seamless. If you’re interested in the future of HCI and vSAN, or even considering running container based workloads (orchestrated by Kubernetes perhaps) on top of vSAN, then this should be a very interesting session for you.
Some other notable sessions that I think will have some very interesting content.
Deep dive: The value of running Kubernetes on vSphere with Frank Denneman and Michael Gasch [CNA1553BU] – Monday, Aug 27, 3:30 – 4:30 p.m.
This is another session that I don’t want to miss. It ties in very closely to the session that I am delivering with Christos, but this one is more focused on the vSphere platform rather than the HCI/vSAN platform. There are still many common themes running through both sessions, and again, if you are already running K8s or plan to run K8s on top of vSphere, check this session out. Frank has done an excellent write-up here. And even Google see the value of running K8s on-prem on vSphere, with their GKE On-Prem alpha announcement at GCP Next 2018 earlier this week. I already got a sneak preview from Frank, and this will be a great session!
A deep dive into why storage matters in a Cloud Native world with Myles Gray and Tushar Thole [HCI1813BU] – Wednesday, Aug 29, 8:00 – 9:00 a.m.
Tushar is the guy who led the effort on Project Hatchway, the ability to create persistent storage for containers in docker, Kubernetes and even PKS (Pivotal Container Service) along with policies and data services on vSphere storage. While containers themselves may be ephemeral, the data needs to be stored somewhere, right? Along with Myles, I’m certain that this session will highlight lots of cool persistent storage examples, and the power of hatchway to dynamically create persistent volumes with appropriate data services and policies as needed. This is certainly a session I want to get along to see.
Best practices for deploying DataStax workloads using vSAN Host Affinity with Caixue Lin and Kathryn Erickson (DataStax) [HCI2001BU] – Wednesday, Aug 29, 12:30 – 1:30 p.m.
Datastax focuses on delivering custom-built Cassandra databases. Working with our vSAN engineering teams, they have developed the optimal method for deploying their workloads on vSAN. Like I mentioned previously, we are seeing many customers deploy next-gen type applications on vSAN, and Cassandra is one of those we keep coming across. So I’m definitely interested in learning more about this, and if you are a customer already using or looking to use Cassandra on vSAN, then this could an interesting session for sure.
That’s it from me – hope to see you there.