1. Home
  2. Server Virtualization Software
  3. VMware ESXi Reviews
  4. User Review of VMware ESXi
De facto Virtualization Standard
February 20, 2020

De facto Virtualization Standard

Anonymous | TrustRadius Reviewer
Score 10 out of 10
Vetted Review
Verified User

Overall Satisfaction with VMware ESXi

At ~98% virtualization, VMware ESXi forms the core of our server infrastructure. It's an exceptionally mature product that grows with organizational needs. In our organization it is used in two separate AD domains, each instance managed separately. One of the key features of ESXi is high availability; virtualization prevents a single hardware failure from resulting in service loss. If the hardware running ESXi fails or develops a fault condition, the virtualized server can be automatically or manually migrated to a functioning host while the fault condition is addressed. An additional layer of resilience is provided by snapshot technology; prior to making any system changes a snapshot can be taken to supplement existing backups and provide a point-in-time recovery. Snapshots also make backups less intrusive when a compatible platform is used.
  • High availability/vMotion for hosted VMs is a must-have in any business. No one wants their systems to fail, knowing that the hosted systems are protected and always available is a great stress reliever.
  • Load and resource balancing (with proper licensing) keeps busy servers from consuming all available resources on a given host. DRS and StorageDRS make short work of balancing workloads, I find this a must-have feature.
  • There are some odd issues with VMware's virtualized network drive (VMXNET3). On occasion, after a reboot of a Windows-based VM the NIC will fail to bind properly and network access is unavailable until an admin intervenes by disabling/re-enabling the adapter. While it's possible that our environment is a contributing factor, this never happens on VMs using Intel E1000 emulation, only the paravirtual NICs.
  • Logging is extensive but difficult to work with. VMware's solution is a product called Log Insight, which comes at additional cost. Fortunately this is somewhat mitigated by the extensive support documentation and robust user community, but in the heat of the moment obtaining the required detail can be a trying experience.
  • We have successfully run many workloads in ESXi, from backup systems to Exchange servers. Hardware costs are predictable and easier to manage, and ESXi hosts can be introduced or retired with little to no impact to production systems, allowing for rolling hardware upgrades.
  • Management time of individual staff is reduced. This is enabled by the powerful and flexible automation options in the platform.
Hyper-V is a viable solution, but in my experience the management toolkit has always been its Achilles heel. vCenter provides everything in one UI, whereas Hyper-V requires multiple tools to accomplish similar tasks. Licensing costs are higher with VMware, mostly because they can't just bake it into their operating system, but with the added cost comes top-notch support via phone or email. Hyper-V is perfectly capable, but doesn't enjoy the same wide base of community support that VMware does. ESXi is also extensible by third parties to a degree unmatched by other platforms. There's a reason many organizations feel that the ROI is worth it and invest in VMware, making it a de facto standard of sorts.

Do you think VMware ESXi delivers good value for the price?

Yes

Are you happy with VMware ESXi's feature set?

Yes

Did VMware ESXi live up to sales and marketing promises?

Yes

Did implementation of VMware ESXi go as expected?

Yes

Would you buy VMware ESXi again?

Yes

ESXi excels at providing stable, highly-available resources for most applications. The vast majority of workloads can be virtualized without a penalty. Hardware accelerated assistance is exposed and effective, and the options are plentiful. Adding memory or compute resources is dead simple. Pooling of resources and defining resource priority is top notch. A key to succeeding with ESXi is to give it more than it needs - more memory, more CPU, more network bandwidth. ESXi will manage those resources effectively so long as you give it room to work.

However, with high-transaction databases and some latency-sensitive applications, ESXi may not be the best solution. I've encountered long snapshot times that have disrupted SQL operations when I/O is frozen to create the snapshot. VMware has documented this and actually recommends avoiding snapshot technology in these situations. While not a common scenario this can result in unexpected production issues without thorough, careful planning. ESXi is also not suitable for devices or environments where outboard hardware (USB and other devices) are required. Lastly, the floppy drive has been dead for years. Why is it still present in the default hardware for a virtual machine in the year 2020? Does anyone actually have a need for floppy emulation?

VMware ESXi Feature Ratings

Server Virtualization
10.0
Virtual machine automated provisioning
10
Management console
10
Live virtual machine backup
10
Live virtual machine migration
10
Hypervisor-level security
10

VMware ESXi Support

I can't say enough good about VMware's support team. To an individual they take ownership of the case, provide thorough answers, and follow up regularly. On one occasion, a problem we experienced with NSX Endpoint was escalated to development for a permanent resolution after a workaround was found. In my experience, most companies would have tried to find a way to close a case like that instead of taking it all the way. Most importantly, when production is down and every second counts, they VMware teams understand that urgency and treat your issue as if it were the only one they had to deal with. You can't ask for better.
ProsCons
Quick Resolution
Good followup
Knowledgeable team
No escalation required
Immediate help available
Support understands my problem
Support cares about my success
Quick Initial Response
Problems left unsolved
Yes
We run a 24x7x365 production operation. Knowing that someone will pick up the phone and troubleshoot a severity 1 issue is well worth the price.
Yes - In fairness the bug was with an add-in component (NSX Endpoint, formerly vShield). Nonetheless, support owned this, provided a workaround, and ultimately provided a release that addressed the initial problem.
I've previously noted that we encountered an issue with NSX Endpoint. This manifested itself as a problem in networking on individual servers - mail delivery was slow, for example - and VMware treated this with high priority. The tech walked me through a method of managing the issue we faced while the development team addressed it. The tech even followed up a month later to see if we were still facing the same problem and to ask if there was anything additional he could offer by way of support.