Cisco UCS Manager Room for Improvement
Infrastructure Architect at a transportation company with 1,001-5,000 employees
So far the only challenge we face with Cisco UCS is during firmware upgrades. If it happens that there is a failover, and we need to change something in the system, this is where we can run into problems. We can't upgrade the firmware for each component one-at-a-time. It is not a method that will work in a practical way in a larger or global network.
Nowadays it is some sort of a status symbol or a business necessity for a customer to be in various geographical locations. Because the client can have locations in Australia and the U.S. — in different regions of the world — that tends to make the maintenance of the firmware more difficult. The various business locations offer challenges in that way.
Usually, when we procure the blades, everything has the same firmware level. This makes sense and is fine if installing in a singular location and for new installations. Everything will match. If it is a new installation and the hardware was not procured at the same time, the firmware for all the components can easily be upgraded because it is still before the implementation.
But later — say after one year — a customer needs to expand. If we are procuring a new blade, the new blade will come with the new firmware. When the new blade is mounted into the chassis, the old alignment will not understand the new blade because it has new firmware.
In that case, you need to downgrade the firmware for the new blade or upgrade the firmware for the entire environment. During the firmware upgrades, we would definitely need to take downtime in some cases and the downtime would take too long. We face that challenge all the time in having to choose which path to take during the upgrade. But because of the obvious issues with upgrading the entire environment, it often looks like a better solution to just downgrade the one new blade. We need the option to downgrade or choose the firmware for the component because we cannot upgrade the entire environment. In many cases, we cannot take the downtime for the entire environment because of what it means to the network and the business.
We should have chances to work with firmware levels in one or two firmware versions and it should be easier to do. Everyone would be comfortable with that. Otherwise, in some cases, there is no point in providing a new blade. Customers will hopefully grow and need new blades. We don't want any extra risk with downtime.
So Cisco should make an improvement in the firmware upgrade process. No one is providing this kind of solution. But if Cisco would improve that firmware issue, that would be great.
A new feature that I would suggest is to have the possibility of different types of connections. Within the full-width blade, there are two types of blade: full-width and half-width blade. In the full-width blade, when one link fails, the other link will take care of the entire load. The half-width blade doesn't have that kind of input. It has only one link. If one link goes down, the entire blade goes down. So Cisco should include the feature like that in the half-width blade so it functions more like the full-width blade and is not prone to failure.View full review »
If there were a way to make the upgrade process better, it would be great. Our most important issue right now is to resolve the installation barrier so we can use the product features fully. We needed to phase in new hardware and try again with Cisco HyperFlex M5.
Also if we had one interface, it would solve some operating issues. If we can control all devices from one interface, I think it would help a lot.
IT Infrastructure Architect at a tech services company with 11-50 employees
The interface and the way it is constructed is very complex. They should work to simplify it. It's quite difficult for somebody who doesn't know the product very well. Users should be able to get proficient with it faster. There's definitely room for improvement there.View full review »
Systems Engineer at a hospitality company with 10,001+ employees
In terms of what can be improved, the help dashboard could. Usually, we use vCenter. If I go to an option called Performance Monitoring, it gives us metrics in real-time.
The performance dashboard should be out in front of the UCS Manager - as soon as you log in you should see the dashboard. That should be improved. Another thing is the inventory management dashboard. Inventory is like a configuration database. So we should also be able to pull all the details which can give answers so we do not wonder about the HCI data.
Also, scalability could be improved in this solution.
Lastly, it should be more user-friendly because Cisco is a bit of a complex solution. So we are running the VMware environment and it has added capabilities of management.
Network and Security Manager at a energy/utilities company with 5,001-10,000 employees
Firepower has weaknesses. I had to load several partitions to improve it.
A smart office solution provided us with a demo showing us the camera qualities. I believe that Cisco is moving forward with this.
In the next release, I would like to see improvements made to their security. Also, they have to introduce firewall compatibility in the UCS, as with Firepower where they have a dedicated box with their software, switches, and routing. It's a one-box solution and it would be a huge benefit for Cisco.
Cisco depends on other vendors like IBM and HP for the hardware. Cisco should improve its hardware manufacturing in regards to the UCS and the use of other vendor's hardware.View full review »
Integration can always be improved, but that is not an issue for me because I work in a bank in Turkey. In Turkey, there are some regulations regarding Co//ab in the banking industry. Authorities are not permitted to use Co//ab for banking, so integration is not a use case for us.
The stability could use some improvements.
We have three data centers and if we could manage all three data centers using one interface, it would be great.View full review »
System Specialist at MTN
I found it a bit of a challenge to get training on UCSM. I've been trying to get that for some time now. I feel like I have to figure out a lot of things myself. For years I've to log calls with support whenever I've got challenges that I cannot resolve. If I had some training or more manuals, I'd be better able to handle more things on my own.
They should work to simplify the server creation template.View full review »
IT Infrastructure Engineer with 51-200 employees
The automation within the solution needs to be simplified.
In the next release, the solution should have a central view center or offer administration capabilities for different environments.View full review »
Sr. Lead Consultant - Infrastructure | Virtualization & Cloud Computing | SDN |NFV at a computer software company with 1,001-5,000 employees
I would like to see Cisco UCS optionally work as a hyper-converged system because right now, it only operates as a converged system.
Adding another layer of embedded virtualization would allow us to sell this as one unit, like Nutanix or VxRail.
Implementation and Support Engineer at a tech services company with 11-50 employees
Its user interface can be improved. It can be more user-friendly.View full review »