4 out of 5.0, Reviewed Jul 24, 2016
Overall been a positive experience, no major outages. If they could improve the design of server housing.Dell by far has the best servers.
Study up on it, play with the simulator, it's a learning curve, being a small shop, changes come few, so it's constantly remembering..
Make sure you get something like a "FlexPod".It helps when dealing with support.
Playing 20 questions with support.Not having basic monitoring for free at the start.
Deploy fewer with more RAM and beefier CPU's.
4 out of 5.0, Reviewed Jul 22, 2016
Performed as expected.
Do your due deligence. Look at all available technologies in this space.
Service profiles and fiber interconnects.
Offer better solution for rackmount form factor.
Evaluate other blade technology, Dell. etc.
Again, service profiles are life saver and some of the out of the box APIs are very valuable.
Sev1 are addressed within SLAs.
Service profiles are great.
4 out of 5.0, Reviewed Jul 21, 2016
Be aware the FI is limited to 24 chassis.
Central admistration and ease of replacement.
Better documentation of service profiles.
Devise a better cabling strategy for continued growth of the UCS product.
4 out of 5.0, Reviewed Jul 21, 2016
We have had very good luck overall with our UCS environment. There have been the occasional hiccups with components failing and some odd inter-vendor incompatibilities but we still maintain that the flexibility and redundancy gained overall is a huge advantage for our physical server and VM environment.
Send your IT Staff to training and work with a solid, reputable integration provider when implementing your UCS environment, no matter how small you might be starting.
The flexibility and management of UCS are amazing. Maintenance windows aren't as daunting due to the ease that service profiles are moved around.
The one dislike we've come across would be the somewhat cryptic messages that crop up during firmware updates on the UCS hardware. Sometimes they refer to a prerequisite in a roundabout way, sometimes they refer to a hardware issue that isn’t severe enough to log an error but they almost always require a support case to decode.
There are a couple points in our implementation I wish they would have steered us a little more clearly and kept us from digging a hole that we only noticed quite a way down the road, long after our UCS environment was stood up.
We would start with multiple UCS domains from the beginning of the implementation if we could.
There are lapses in communication here and there but overall the support has been good.
We had a couple of decisions that we did end up going back and changing, but there were no huge snags or show stopping issues in our setup.
5 out of 5.0, Reviewed Jun 6, 2016
When we envisioned our new environment "Shared Computing Environment" we brought in leading vendors for compute, storage, and virtualization. Key strategic partners became CISCO, EMC, VMWare and Microsoft. All worked together seamlessly to develop an environment like the CISCO V-Block, but more advanced. We ended up taking 34 equipment racks and placing a greater capacity in the data center reutilizing on 7 racks, realizing a 63% decrease in power utilization and an almost equal reduction in cooling requirements.
Ensure that you have a complete phased implementation plan that includes training for operations and maintenance personnel.
It is a hyper-converged product that is totally integrated.
I actually like everything about it.
High capacity, lightning fast.
The beauty of the environment is that it comes already integrated.
4 out of 5.0, Reviewed Jun 2, 2016
Cisco UCS seems to be a proven leader in the converged infrastructure space. Most of the product seems to be well developed and tested. There are still some bugs that can cause serious issues, but overall I think this is the best solution available.
Start small with a Proof of Concept, but make sure that your Proof Of Concept encompasses all versions of the hardware that you will be interfacing with. Different storage arrays for example, interact far differently with UCS.
When designed correctly the UCS environment can save countless labor hours and allow engineers to focus their time on other issues. The UCS system also helps establish standards which is something that many organizations struggle to do.
When you converge architecture and start using automation, you run the risk of creating big issues. A small error in a process can cause massive outages, configuration issues, etc. For example we ran into a bug that deactivated entire VSANs in our SAN. You need to make sure you understand every step the system is automating for you, so that you are ready to support it when it doesn't work.
Did a thorough check of the environment, hardware and code levels to ensure compatibility.
Spend more time understanding the compatibility matrix, caveats, and current issues. We were bitten a couple of times due to either incompatibility or perceived incompatibility where the support representative and/or documentation was incorrect.
UCS appears to have the ability to directly interface with most products we use and allows the flexibility to create custom commands/tie in for products when not directly supported.
It seems like we often run into issues which are then qualified as bugs with little to no ETA for resolution.