Hi HCX community!

Ufff – it has been entirely too long since my previous update. As you may imagine, times are interesting at the VMware office with all the ongoings (we don’t talk about Bruno~🎶)… and then everywhere else with all the talk of UFOs and aliens?, and then an infernal summer season that doesn’t stop dishing out 3 digit days (at least in the beautiful beast that is Texas). In any case … I was fortunate to be able to squeeze in a much needed summer getaway: a 4000+ mile Americana roadtrip with my favorites!

I digress ..

What is happening in the HCX world

Today the HCX product line exists as three products with distinct key dates, markets and capabilities (with core feature overlap). For clarity’s sake I want to quickly clarify them:

1. VMware HCX:
This is the HCX most of our production customers know.
– We release new functionality every 3-6 months in minor releases (e.g. 4.5 4.6), and support each of these for one year. Maintenance releases (e.g. 4.5.2,4.6.1) provide fixes and have no bearing on the minor version’s key lifecycle dates.
– Functionality in two tiers:
HCX Advanced (Bulk Migration, Serial HCX vMotion, WAN Opt, L2 Extension)
HCX Enterprise (Replication Assisted vMotion, Mobility Optimized Networking, OSAM, Network Extension HA, Migration Estimation, Mobility Groups).
– Functionality is orchestrated per site, on the HCX Managers.

2. HCX+
This is the newest member of the HCX family. The SaaS-powered HCX.
– HCX+ is in its Initial Availability. I have had the pleasure to work very closely with
– HCX+ follows a more agile release cadence, with the VMware hosted elements receiving continuous improvements, minimizing site level upgrades.
– HCX+ customers have access to the HCX Advanced/Enterprise tier features and also new capabilities developed and workflows redesigned exclusively in HCX+ (where we have had the flexibility to take our learnings and rebuild the user experience). HCX+ is where the bleeding edge HCX experience is had.
– Functionality is centrally orchestrated: Users access the HCX service and control migration securely and with minimal hassle by browsing hcxplus.vmware.com.

3. HCX for AWS GovCloud
– This is the version of HCX that is deployed to VMware Cloud on AWS for GovCloud, where agencies require systems to be FedRamp High certified.
– HCX for GovCloud must specifically run versions 4.3.9 and 4.5.1, and must observe the lifecycle dates related to these versions (the HCX 4.7 update below is not applicable for GovCloud deployments).

These are the 3 products – what’s new in HCX 4.7 (in the update below) does not universally apply to HCX+ or HCX for GovCloud and vice-versa.

With that clarification out of the way let me get into the update.

What is New in HCX 4.7

At the end of July, we released VMware HCX 4.7! This new minor release has updates three key areas :

1. HCX for NSX V2T Migration
   
2. Migration Scale (RAV/BULK)
   
3. Interoperability:
   • HCX compatibility with vSAN 8 Express Storage Architecture
   • HCX for NSX 4.1.1 & 3.2.3
     

1. HCX 4.7 for NSX V2T Migration (NSX FEDERATION)

With HCX 4.5 we released Workload Migration as integrated functionality for NSX Migration Coordinator. See the blog here. The 4.5 integration allowed customers to perform a lift-and-shift style NSXv to NSXT migration with the new User Defined Topology and Edge Migration Mode. This mode allows NSX Migration Coordinator to establish a direct L2 bridge between NSXv and NSXt Edge hosts, migrate the NSX configurations and use HCX to orchestrate direct vMotion (host only or host+storage) without reducing the security posture of the workloads during the migration. In HCX 4.5 version, support was for cluster to cluster projects within a single vCenter Server. In HCX 4.6, we expanded to support a similar NSX Migration Coordinator project across vCenter Servers (a source VC with NSXv and destination VC with NSX-T).

The latest updates in HCX 4.7 expand the integration to include migrations from Cross-vCenter NSX (NSXv) environments into new NSX Federation environments. In this mode Migration Coordinator provides the L2 bridging, NSX configuration conversion and object preparation and the HCX V2T mode allows direct cross-vCenter vMotion (host only or host+storage). The traditional HCX migration modes (e,g, Cold, HCX-vMo, Bulk,RAV) are also available for migrations into Federated environments. (note that HCX provides workload migration and not provide L2 connectivity into NSX Federated destination).

2. 2X Migration Scale (Concurrency) Improvements

I have been championing… maybe hooting and hollering about the critical need to transform HCX to scale comfortably in an industry marching towards 200/400Gbps (where 100Gbps is gaining a foothold). To that effect, team HCX is planning multiple improvements under one performance and scale initiative. It is exciting to see the first deliverable in HCX.

In HCX 4.7, we’ve nailed the changes required to squeeze 2X improvement in the HCX management plane. This allows a manager to comfortably manage (an officially supported) 600 concurrent migrations across all it’s site pairs/service meshes.
For those that live in the danger zone, we observed a manager uncomfortably handle 3X+ replication (RAV only) load, but that requires carefully managing things.

The bandwidth-tiered guidance in the HCX Network Underlay Characterization and Performance Outcomes holds true, since that is focused on HCX’s data plane limits.

This scale improvement requires HCX Manager VM configuration changes and some configuration tuning. For those details please check out VMware KB 93605.

Beyond excited for every upcoming capability on this front.

3. Interoperability Update

HCX is now compatible with NSX 4.1.1 and 3.2.3; this is our basic interop ensuring core services are available on the latest VMware software. The following benefits apply with HCX 4.7:

• HCX migrated virtual machines can be connected into NSX 4.1.1 or 3.2.3 segments.
• HCX can be used to extend VDS/NSX networks as NSX 4.1.1 or 3.2.3 segments.
• NSX 4.1.1 or 3.2.3 segments at the source can be extended.

HCX 4.7 adds compatibility for migrations to environments built on vSphere 8 with VSAN 8 Express Storage Architecture (ESA). Some ESA highlights:

• The space efficiency of RAID-5/6 erasure coding with the performance of RAID-1 mirroring.
• Adaptive RAID-5 erasure coding for guaranteed space savings on clusters with as few as 3 hosts.
• Storage policy-based data compression offers up to 4x better compression ratios per 4KB data block than the vSAN OSA.
• Encryption that secures data in-flight and at rest with minimal overhead.
• Adaptive network traffic shaping to ensure VM performance is maintained during resynchronizations.
• The lower total cost of ownership (TCO) by removing dedicated cache devices with the flexible architecture using a single tier in vSAN 8.
• Simplified management, and smaller failure domains by removing the construct of disk groups.
• New native scalable snapshots deliver extremely fast and consistent performance.  Achieve backups more quickly through snapshot consolidations that are up to 100x faster.

You can find more info here: VSAN 8’s Express Storage Architecture

⚠️ Heads up – It is absolutely critical to plan HCX 4.7 with ESA migration projects! HCX 4.6.3 will block migrations to VSAN 8 ESA destinations, versions 4.6.2 and older will result in migrations that need recovery steps. See VMware KB 92929.

As a part of this interop update, the Host Based Replication service has been updated to version 8.7 (from 8.4) subsuming many fixes.

In closing

Time really got away from me writing this one. I’ll be back again within a week with VMware Explore related ongoings and updates on the HCX+ front. For now I will say I hope to see you there! I will definitely be in Las Vegas a couple of sessions and many scheduled meetings. If you are in one of these I look forward to our time there!

—
✌🏼👽 Gabe

Summer roadtrip 2023 Memories