What is our primary use case?
We currently utilize the Kemp LoadMaster VLM-500 series virtual appliances in our VMware environment. Initially, we were running with only a single appliance, but recently added a second to allow for High Availability pairing between them.
Our primary use for the LoadMaster appliances is to provide load balancing between our two Microsoft Exchange servers in our Office 365 hybrid environment. This has allowed us to obtain redundancy between them, rather than having a single point of failure for our email environment.
How has it helped my organization?
Before we implemented the Kemp LoadMasters into our Microsoft Exchange environment, we were operating our SMTP relay and OWA traffic directly to the Exchange server itself. This put us in a bad spot when it came to patching and maintenance, as well as when we had service-impacting Exchange issues occur on the server.
By implementing load balancing via our Kemp LoadMaster, we were able to add an additional Exchange server into our environment, which allowed us to failover manually for maintenance, as well as have a backup for unexpected issues
What is most valuable?
The most valuable feature so far has been the high-availability options that allowed us to add an additional Kemp LoadMaster VLM virtual appliance into our VMware vSphere environment, to provide failover for our existing LoadMaster. We have never run into any service-impacting issues with the appliance itself, but having a second one allows us to reduce the downtime associated with maintenance upgrades to zero.
I also appreciate the online licensing feature, which lets us update and install licenses on the appliances by simply logging into our KempID account. It was nice not to have to fuss with licensing and key file download/uploads.
What needs improvement?
So far, the only hitch we have run into is that would have been nice to have an easier method to add allow/whitelist entries into the Access Control lists for virtual services.
Following an upgrade, we inadvertently lost all of our Access Control whitelist entries on one of our virtual services. Thankfully we had a backup of them, but to plug them back in, we had to enter them all manually. This ended up being a bit of a pain.
It would be great if there was a way to upload a .CSV file of ACL entries into the access control list, rather than having to add them one at a time.
For how long have I used the solution?
We've been using Kemp LoadMaster for approximately five years.
What do I think about the stability of the solution?
We ran a single (non-redundant) LoadMaster appliance in our VMware environment for close to five years with zero outages caused by it, so I am very impressed with the stability of KEMP's devices. The only reason we explored adding a second appliance for High Availability was to make maintenance easier.
We've had other virtual appliances in the past that had to be rebooted occasionally, but that is not the case with the Kemp LoadMaster. They are rock solid in terms of stability.
What do I think about the scalability of the solution?
We so far have not come anywhere near utilizing the total throughput of the appliances we have. They have no problems keeping up.
How are customer service and technical support?
The customer service was great! They helped me to quickly resolve two licensing-related issues with our additional appliance purchase.
Which solution did I use previously and why did I switch?
We did not previously use load balancing, Kemp was our first choice.
How was the initial setup?
The initial setup is a breeze and is well documented.
What about the implementation team?
It was implemented in-house using only Kemp documentation and no outside support.
What's my experience with pricing, setup cost, and licensing?
Pricing for the perpetual licensing was fair to us for the features and ease of use we received.
Which other solutions did I evaluate?
We also evaluated Citrix NetScaler.
What other advice do I have?
The VLM appliances boot insanely quickly, which is really impressive. During power cycles, we only drop between two and three pings to the devices.
Which deployment model are you using for this solution?