In this third blog in the Project Ensemble Deep Dive series, I will take you on a tour of the new user interface in development. In the first blog I discussed the API first approach for Project Ensemble and in the second blog you learned about the platform architecture. In both blogs, I discussed the federated data model used by Project Ensemble to collect and rationalize cloud management inventory and configuration data into a single, extensible Entity Data Service, upon which other services and capabilities can be added. In this blog you will see how this federated view is surfaced to the user.
Before we begin, please remember that anything discussed here is subject to change and may not make it into the final service. With that I will share our legal disclaimer.
“This blog may contain product features or functionality that are currently under development. This overview of new technology represents no commitment from VMware to deliver these features in any generally available product. Features are subject to change, and must not be included in contracts, purchase orders, or sales agreements of any kind. Technical feasibility and market demand will affect final delivery. Pricing and packaging for any new features/functionality/technology discussed or presented, have not been determined.”
Information Overload
A common problem faced by users of any management console is how to find the important information quickly and then act on that information. I personally recall my first experience with many IT tool’s user interface just spending time clicking around and trying to discover what information was available to me and what I could do with that information. Some user interfaces do a better job than others and the experience can differ vastly from product to product.
Considering that most IT user interfaces are designed for a trained or experienced practitioner, it is not surprising that many first-time users feel lost and spend much of their time just learning to navigate and explore. Compound that by the number of tools that a practitioner uses daily. Introducing a new tool, interface or service increases the difficulty in getting to the information needed to perform common, daily tasks.
When I was a system administrator, I had many lengthy … let us just say “discussions” with application owners, developers and database administrators explaining why I would not grant them access to the IT tools I used. While I was sympathetic to their desire to perform routine tasks and understand the health and performance characteristics of their workloads, the thought of giving users access to my tools sent a cold chill down my spine. Besides, the tools were designed for system administrators, network administrators, storage administrators and not the needs of developers and application administrators. They are expert-focused, and purpose built for siloed IT.
Now days, the developers finally do have access to the tools IT typically restricted them from in the past. Thanks to public cloud services, developers are free to deploy and manage their own platforms and even infrastructure as a service without the need to wait on IT. The trouble is, IT has lost governance and control for security, performance, capacity, and cost. The problem has essentially flipped. Controlling and managing resources, users, applications, workloads and policies across multiple clouds is practically impossible to do without a common, federated view.
What would a singular user interface look like with the following capabilities:
- Customized visualization of a user’s applications in a way that intuitively shows the information they care about most.
- Consolidation of IT tool data into a higher-level view with the ability to launch, in context, into specialized tools and services.
- Global search capabilities across all environments and clouds.
- Self-service for consumers from cradle to grave for their applications and workloads.
These features would finally address an age-old problem most organizations face, which in essence boils down to the unification of IT and business to accelerate growth, adopt new technologies and mature operations at a scale not possible previously.
Empowering a Cloud Operating Model
If you are not familiar with the concept of a Cloud Operating Model, you can read more about it here. To summarize, a Cloud Operating Model allows customers to move from a traditional IT operating model and accelerate digital transformation in four key areas:
- Automated Service Delivery to increase speed and agility
- Performance Management to optimize and drive efficient operations
- Cost and Capacity Management to govern and optimize spend and drive accountability
- Security to manage risk and compliance
This means customers will need insights to many datapoints for their applications as well as day-to-day lifecycle management. This is all possible within Project Ensemble’s user interface, which is both persona and application centric in presentation of the information required to adopt a Cloud Operating Model.
Persona and Application Centric
Two key principles guide the Project Ensemble user interface: Personalized user experience and application-focused views and analysis. This means that both cloud consumers such as developers and application owners and cloud providers such as cloud administrators and network engineers can all use Project Ensemble in a way that is meaningful to each of them. And it further means that the most important part of the cloud, the applications which run on them, are the focus for everyone.
Upon login to the Project Ensemble user interface, it is easy to see that this is not your typical, legacy IT view. Instead of focusing on infrastructure, cloud resources or networks, a user is presented with all the applications associated with project membership. Right away, they can see their applications and the workloads that comprise them.
Applications are discovered by vRealize services, such as vRealize Network Insight Cloud, and are curated in Project Ensemble. Relationships between applications and application components are also derived from the underlying services feeding Project Ensemble.
Note the additional details, actions, and insights for the selected application on the right side of the screen. This provides a consolidated view from all vRealize services feeding Project Ensemble which reduces “mouse click fatigue” for the user. All practical information is exposed within a single view.
From this view, the user can click into applications for more detail.
From here, the application tiers and tier membership can be explored. Again, these tiers are determined by data provided from services feeding Project Ensemble and with additional insights from Project Ensemble. The application owner can also further curate these tiers and add additional members or tiers as appropriate.
Now let us consider the perspective of an IT practitioner. Cloud providers are most concerned with making sure the infrastructure and services are meeting SLAs for performance, availability, and security. Instead of top-down view, they are more likely to start in a view of the infrastructure. In this screen shot we can see such a view.
The view is consistent with the application view, and functions similarly drawing attention to problem areas, providing insights and available actions as well as summary information from vRealize services for each infrastructure entity. Of course, this view does allow the user to navigate to related applications. This is helpful to determine the potential impact of an infrastructure outage or problem on the hosted application workloads. Recall that Project Ensemble knows which projects own which applications, so alerting application owners to potential issues is easy.
Of course, the cloud provider persona also needs to respond to application owner’s requests for help. Navigating to the application in question directly to start investigation of issues top-down is an important capability. For that, Project Ensemble offers a global search, allowing users to enter a free text search for applications and other entities.
As shown in the screen shot above, the search function gives fast results with an auto-suggestion so that users do not have to complete a query to get results.
I have mention “insights” several times in this blog post, and you might be wondering what that means in Project Ensemble. Simply put, insights help the user understand the problem and the context of the problem as well as likely outcomes if action is not taken.
Insights are based on observations, which aggregate alerts, notifications, and other notable changes from the data feed from the underlying vRealize services.
For example, the insight shown below describes a memory pressure threshold violation on a VMware Cloud on AWS cluster, which will result in an Elastic DRS deployment of an additional host.
Included are the high-level details of the problem, the timeframe for any projected impact and the cost for adding an additional host. Based on this, the user can take any of the actions presented to either reduce the memory pressure and avoid the automated scale-out, or decide to increase capacity in the cluster right away.
Not shown in the screen shot above, but presented below, is an overview of the evidence gathered by Project Ensemble to support the observation and insight.
You can see that the capacity forecast has predicted this eDRS action based on the eDRS policy settings, historic memory usage and trends. Also provided are details on top consumers of memory, with optional details via a breakdown of resource usage for each virtual machine.
Simplifying and Unifying Cloud Management
Keep in mind, as previously stated, that these screen shots and workflows are currently in development and may change significantly before availability. However, I believe this will give you a solid understanding of the purpose and intent of the Project Ensemble user interface. I hope you will also agree that the objective of Project Ensemble, to unify and simply multi-cloud management for both cloud consumers and cloud providers, is met by the design and principles of the user interface.
In this blog series you have learned about our API First approach using GraphQL to elegantly query Project Ensemble and all vRealize Services from a single API request. You also saw how Project Ensemble architecture is a foundational approach for a common data model for existing and new services from VMware. And of course, this blog explored the Project Ensemble user interface which focuses on persona and application-centric views and workflows.
Want to Learn More About Project Ensemble?
If you are interested in being considered for the beta to provide VMware with input, please let us know here (https://www.vmware.com/learn/1243700_REG.html) and your eligibility will be evaluated.
