vSAN Hyperconverged Infrastructure Virtual Volumes (vVols)

VMworld Read-Out – A Whirlwind Review of the Best Attended Technical Storage Sessions

With over 100 storage-related sessions at VMworld 2017, it was all but impossible to see them all (though we’re sure you tried!).  Fear not, we’ve gathered a run-down of the most popular technical sessions you may have missed.

The most popular single storage tech session was “VMware Cloud on AWS: Storage Deep Dive”, which shouldn’t be a surprise given the overwhelming interest in this new on-demand service. The primary goal of the service is to provide a consistent VMware based workload and management framework for all applications on the Amazon public cloud coupled with taking care of the operation of the software defined datacenter infrastructure (vSphere, vCenter, NSX and vSAN) so that our customers can focus on deploying and running their workloads.

The initial offering of VMware Cloud (VMC) on AWS is geared towards the highest availability and performance, with eight 1.8TB SSDs per node, arranged as two disk groups with one cache and three capacity drives per group.  Cluster sizes range from four to sixteen nodes in a single AWS Availability Zone.  The offering provides flexibility for RAID & mirroring options. This solution is a “no-brainer” for organizations who love our software-defined datacenter but also want a public cloud deployment option. Hear it for yourself here.

For practical advice and “this is the way to do vSAN right,” two no-nonsense system engineers (Joshua Fidel and Ron Scott-Adams) presented “Best Practices for vSAN Design: A Real-World Perspective.” This session covered nearly every aspect of a vSAN deployment and included real world data that doesn’t make it into the product documentation. For example, actual CPU consumption and IOPS rates in production.  If you’re ready to deploy vSAN or just want to make sure you’re on the right track, this is the session for you. Watch the replay here.

There were multiple “What’s New in vSAN” sessions for those already familiar with the product. One clear take-away was the amount of robust, enterprise grade features, that were added to vSAN in 2017.  In a recent Tech Validate survey of vSAN customers, more than 60% of respondents said they are using vSAN for business-critical applications. High availability and performance are key for those mission critical applications and the sessions took a deep dive into the many, recent vSAN enhancements in that area.  “A Closer Look at vSAN Networking Design and Configuration Considerations” was another top performer among VMworld audiences. It provided concrete advice on how to architect your network to support vSAN, and closed with some eye-popping test results showing just how important low latency and packet loss are to implementing hyper-converged infrastructure. View it first-hand here, and take notes!

Virtual Volumes (vVols) sessions were also popular, holding two of the top spots.  vVols is an integration framework for VM-aware storage and policy based management.  It’s included in all vSphere editions, and is supported by all the big storage vendors. vVols provides a flexible connection to traditional NFS or SAN based storage using a consistent logical interface.  Eliminating clunky interfaces like LUNs that currently keep compute and storage silo’d and lead to needless overprovisioning and management overhead.  vVols allows customers to define storage policy templates (e.g. RAID, snapshots, QoS, Replication, etc.) that can be applied directly to VMs and support multiple storage vendors transparently. The bottom line is this: If you are looking to do hyper-converged infrastructure, then vSAN is your best choice.  If not, then you need to take a hard look at moving to vVols to simplify management of third party storage, and to better integrate that storage with your workloads.

Rounding out the Top 10 sessions was “Architecting Site Recovery Manager to Meet Your Recovery Goals”. This session included a great deal of information on vSAN integration and how to use NSX to really simplify disaster recovery testing.  GS Khalsa closed the session with a demo that showed just how quickly SRM can recover workloads in about 25 minutes for 1,000 VMs! Check out the session and demo here!

Take a look at all of the sessions you may have missed at this year’s events here and start making plans to attend in person in 2018! Make sure to log into your My VMworld account to easily access the session recordings.