On we roll to the final part 3 blog on UCP Advisor v1.2. I jest with the part "Tree" as I was reminded that those unaccustomed to listening to the Irish accent don't know we tend to drop the "h" which sends my kids into hysterics when I say "3".
In part 2 , I covered the essential networking mgmt, some of the day 0-90 administration aspects and integrated data protection management features. I am going to conclude and cover firmware management, some of the bare metal support, integrated operational analysis with Log Insight content pack and close on some of the cloud automation integration aspects using Powershell and vRO.
When it comes to software firmware management, it's one of those areas that normally infuses dread into administrators. We all share the same apprehension when deciding to update some frequently used software package on our tablet/streaming TV that we tend to rely on after poor previous experience. The planning and execution to successfully navigate software upgrades in an enterprise environment of inter-dependent infrastructure components running production workloads, with the mandate to keep the business online, takes it to another time consuming level. Part of the issue is the passage of time between each upgrade requiring generalist admins to re-research and figure out the esoteric operations for each of the individual components based on latest best practices/software versions. With UCP Advisor, we tackled the problem of ensuring the software could update administrators when a suitable collection of firmware was available, give administrators a simple menu of to select where and what could be deployed. This includes BMC, BIOS and BMC firmware upgrades on the compute nodes supporting hyper converged UCP HC and converged UCP 2000 and also the network switches providing either FC and Ethernet services. UCP Advisor will apply the software updates carefully across the range of infrastructure components while ensuring non-impact to services that UCP is serving. It ensures, for example, firmware updates are completed on side A of FC fabric/spine-leaf infrastructure before doing side B, it ensures each or a set of compute nodes is back and operational before progressing to next one. In version 1.2, you can optionally add additional capacity before doing upgrades to ensure sufficient capacity to handle workloads during upgrade cycle. Click the visual below to get some some indication of the administration process.
I mentioned bare-metal. Essentially, UCP Advisor is not restricted to providing infrastructure services to ESXi hosts. With right credentials, UCP Advisor exposes physical infrastructure management so Advisor administrator could automate the creation and hostgroup presentation of iSCSI LUNs to a RedHat physical node. Other areas covered include storage pools and subset of replication pair management
When it comes to operational insight, VMware's log insight (vRLI) provides a intuitive real time log analysis toolset to provide real time answers and insights to problems or potential operational problems related to systems and services on UCP platform. UCP Advisor comes with integrated content pack which provides the connection to logs from compute, network and storage. In a previous blog, I covered some examples of how a version of this content pack can be leveraged including automatic alerts based on unknown intrusions into infrastructure management domain from suspect IP addresses based purely on real time log analysis. Huge potential to exploit this for many different use cases. Click visual to get a perspective on this. I'm really interested to hear how you are exploiting the log insight content packs.
Finally, UCP and UCP Advisor can be integrated into cloud automation management toolset and processes such as vRealize Automation or similar products like Hitachi Enterprise Cloud. UCP Advisor comes with extended PowerCLI cmdlets and vRealize Orchestrator (vRO) workflows for the majority of the tasks that I've mentioned previously. There is a mixture of foundational workflows and high level workflows that are provided, whether for example its to allow service catalog based creation/deployment of ESXi/datastore infrastructure resources or allow self service VM recovery for tenant users.
To borrow from my "tree" opening, I highly recommend evaluating the infrastructure automation simplicity that UCP Advisor (download) brings to your virtualization/private cloud infrastructure projects so you can manage the forest and not have to climb each and every tree. Stay tuned as team is busy working on next-gen UCP Advisor with even more automation brilliance.