In addition to the record-setting Vegas heat at VMworld 2017, this was a show chock full of surprises. The ServerCentral Managed Services team was able to sneak away from supporting our amazing customers to attend the show this year. We even found time to play a few slot machines and blackjack. Shh…don’t tell our bosses. Now that we’ve had time to recover from the desert heat and get in some last-minute grilling before Chicago winter hits, let’s recap some of the show’s sights and sounds.
Breakout Sessions Galore
Let’s start with the breakout sessions. VMworld boasted over 500 breakout sessions, which was an increase over previous years. It was tough to build a schedule, as there were so many interesting topics to choose from, like learning more about vSAN 6.6, how other companies leverage NSX, and advanced troubleshooting and performance techniques. It was difficult to find time to visit the Solutions Showcase, as I found myself in back-to-back breakout sessions more often than not. This is not necessarily a bad thing!
Show Highlights and Trends
NSX and VMware Cloud on AWS (hereafter referred to as VWC on AWS) dominated the minds of attendees.
I find that the Hands-On Labs and Self Place Labs are good, general barometers indicating what attendees are most interested in during that year’s show. NSX was well represented on the Lab Scoreboard, as was VWC on AWS. I’m recalling from memory here, but nearly the entire leaderboard was either NSX or AWS labs. The vRealize Operations labs also made an appearance.
With a conference as diverse as VMworld, it’s difficult to pick just one thing that dominated the show. That being said, there were certainly a few crowd favorites:
VMware has long established their foothold and now dominance of the enterprise hypervisor market. Now they are turning their sights to Software Defined Networking (SDN) and NSX specifically. As mentioned previously, NSX dominated the lab sessions and was well-represented in the breakout sessions.
Here are some examples from the breakout sessions:
- NSX Advanced Security
- Disaster Recovery Solutions with NSX
- NSX Performance Deep Dive
- Achieving Hybrid Cloud Data Agility Securely with VMware NSX
- Deep Dive into Operationalizing NSX for vSphere
I could keep going, but I think you see the pattern. It's clear that this is a high interest area for VMware and customers alike.
SDN is poised to be another growth area for VMware. More and more enterprise customers are examining what infrastructure can be virtualized beyond bare-metal servers and traditional workloads.
The natural evolution is to look at storage and the network as the next pieces to virtualize. This is not a trend to ignore, as evidenced by VMware’s announcement of a billion-dollar run rate of NSX in this past year.
We found these sessions fascinating. They provided a wealth of material for us to consider, as we are always looking for ways to enhance our managed service offerings. Chief among these considerations are the implications for disaster recovery (DR) engagements and securing East-West local VLAN traffic, otherwise known as micro-segmentation.
NSX for disaster recovery engagements
For the DR piece, NSX provides mechanisms to ensure security policies are replicated between the primary and disaster recovery sites. This is a great benefit to service providers because it’s traditionally a manual process (an administrator commits the policies on the primary environment’s firewall cluster, and then has to ensure they’ve manually committed all required policies on the DR environment’s firewall cluster). This exposes the customer and the service provider to possible human errors.
NSX for securing East-West local VLAN traffic
NSX provides a means to secure East-West traffic without having to implement in-guest firewalls, and then managing those firewall rules and policies individually. NSX brings this up to the hypervisor layer, where you can utilize vSphere tags and other features within vCenter and NSX Manager to create powerful rule sets that apply to existing virtual machines and future VMs automatically. This makes it much easier to keep your risk and security compliance departments happy and pass your audits. Oh, and did I mention NSX can also function as a router, an edge gateway, a load balancer, and an SSL VPN endpoint? You can see why NSX has such intriguing possibilities and appeal to a service provider.
ServerCentral has a long history in the transit and transport network space, so we are well-positioned for this potential transformation. For now, we’re keeping a close eye on how NSX continues to mature and if we can possibly leverage it in the future for our cloud offerings and overall data center network.
Before moving on, I do have to extend a mea culpa to network administrators and engineers alike. Initially, I was naïve and thought that the trend of pushing towards SDN might set us on a path of reducing the importance of our network brethren. My eyes were opened, though, and these sessions really drove the point home that SDN pushes in the opposite direction. Implementing SDN encourages network and cloud administration teams to work closer together, especially when there needs be an understanding of both sides of the equation as cloud computing continues to transform how we manage and deploy workloads in the modern data center.
The vRealize Suite
While not as hip as NSX and VWC on AWS, it was clear that adoption of the vRealize suite was another soft show trend. (By “soft,” I mean vRealize kept appearing as a tool used in many breakout sessions as opposed to an overt announcement like NSX or VWC on AWS.)
In nearly all of the case studies or breakout sessions, one or more modules in the vRealize product suite was used in some capacity. Today, ServerCentral leverages vRealize Operations Manager to manage cloud deployment capacities and advise our customers accordingly as they grow.
While you can achieve some of the same functionality, reporting, and metrics with elbow grease and data gleaned from monitoring and graphing systems, the vRealize Operations Manager simplifies capacity planning, management, and visually representing and reporting on that data.
If you decide to adopt SDN concepts and implement NSX, the trifecta is Operations Manager, Log Insight, and Network Insight. Combined, these remove a lot of the manual processes of understanding what’s going on under the hood with NSX, including troubleshooting, logging, and so on. In essence, it provides the complete picture of your NSX deployment.
Today, ServerCentral already employs our own remote syslog and other services internally to manage our fleet of virtual and physical network infrastructure. As the SDN and NSX trends continue, we’re closely monitoring how this potential shift may impact our continued vRealize adoption should we fully pursue the NSX route.
Automation as a topic isn’t new. As a managed services team, we’ve long heard the siren call of “IT Automation & Orchestration” in various sales presentations from potential vendors. What’s interesting here is to see just how far the concept has permeated the collective minds across the industry in all segments.
It was also interesting to see how other teams use vRealize Automation and vRealize Orchestrator to streamline operations and achieve greater scalability with fewer staff and other resources. This, of course, is one of the holy grails for a service provider.
It appears everyone has to be thinking about automation and orchestration these days, whether they’re a service provider or a single enterprise. At ServerCentral, we’ve built a business around a high-touch approach in designing and providing custom solutions for our clients.
Our teams strive to understand each customer’s immediate and long-term goals and how we can add value versus dropping off a box of preassembled parts and moving onto the next opportunity.
Our challenge has been how to introduce automation and orchestration to improve the onboarding experience and ongoing support for our clients—without losing our ability to be nimble or to provide a customized solution. We’ve been working on a few things behind the scenes here, and VMworld 2017 reaffirmed that we are on the right track.
VMware Cloud on Amazon Web Services
The big one. The crown jewel announcement for the show. VMware Cloud on Amazon Web Services (VWC on AWS). The buzz has been building for some time now and we were all eagerly waiting for details:
- How much will it cost?
- What does the architecture consist of?
- What’s the minimum commit?
- What’s the service level agreement offered?
By now, much of this information is already well circulated, so I won’t focus on that. I do think there were three interesting points to be drawn from the announcement:
First, the initial offering is surprisingly well baked for a 1.0 offering. The user interface (UI) is relatively clean and easy to understand. Given VMware’s history with UI and some of their past choices (such as the flash web client and the vCloud Air interface), this was nice to see. Recently, they’ve made a push to simplify the tenant and consumption experience with vCloud Director, so it is refreshing to see those principles carry over to a new offering such as this. From a technical perspective, there are a few features missing (as to be expected with an initial offering), such as stretch vSAN clusters, deduplication and compression, but VMware has indicated this release favors performance and durability over features. Right now, this service is only available in one availability zone. More availability zones and other features are slated for future releases.
The second interesting point was the focus on NSX-T. This aligns with the push for NSX to become the dominant player in the SDN space. NSX-T provides the network glue between VWC on AWS and your existing on-premise vSphere deployment. It’s clear what VMware has in mind for its end goal and convergence of the traditional enterprise cloud and the public hyper-scale clouds.
The final interesting point is that VWC on AWS is not for everyone—at least not right now.
There’s some technology and features missing. There’s a higher cost of entry with a four-node minimum commitment and failures to tolerate (FTT) needing to be at two nodes at scale. The service level agreements need to improve for larger enterprises to be comfortable running more of their infrastructure on Amazon’s services. Early adopters who would benefit the most are larger organizations that can afford to play in this environment and already had an investment in running applications across VMware and AWS prior to this announcement.
There are some unique possibilities here for the future, and ServerCentral—like everyone else in this space—continues to keep tabs on this announcement closely.
VMworld is always a great conference to attend for us, but this year was definitely one of most memorable with a number of exciting announcements that will have a long-term impact on the enterprise cloud computing landscape.
Our Managed Services team is brimming with ideas and we’re toiling away in the lab working on our next iteration of managed services. And, of course, we are always testing the latest and greatest products from potential new vendors from the Solutions Exchange.
Until next year, my friends!