Earning a Seat at the Table: Hybrid Cloud with Continuous Delivery & Insights (Part 1)
In my previous post, I addressed the need for organizations to transform and adapt their IT capabilities to ensure that they “responsibly move at market speed.” Now, I wanted to talk about how that can be achieved.
There’s a famous quote from Benjamin Franklin, “If I had more time, I would write you a shorter letter” that a mentor of mine transformed into a set of leadership principles that she called the “rule of 3”.
By combining hybrid cloud, continuous delivery and continuous insights, organizations are able to optimize three KPIs:
- Development Velocity: the speed at which app teams can deliver new features to market.
- Shifting operational metrics from hardware-centric (e.g. 99.99% available), to user experience-centric by monitoring, measuring and minimizing failed customer interactions.
- Compliance Response Time: the time is takes to supply evidence to executives, auditors, and regulators to close and audit question.
After setting the vision to “responsibly move at market speed” and defining the KPI’s, you can break the technical organization into three focused, empowered teams:
Team 1: hybrid cloud
Team 2: continuous delivery
Team 3: continuous insights
Each team should be fully empowered to deliver their solution but also held accountable for their impact on each of the three KPI’s. By dividing and empowering, teams are able to avoid groupthink and analysis paralysis, and therefore move much faster; the minus is that you sacrifice some team cohesion, but winning results and a few beers are a great way to bring the team back together at the end. The IT transformation journey can then be split into two phases to better balance near-term goals with the longer-term strategy:
Three principles to govern the technical decisions for delivering hybrid cloud:
- Fewer moving parts in the platform will lead to a simpler, more stable environment. Traditionally, too many teams are required to get anything done, and the ineffective handoff of work between the teams can lead to numerous outages, requiring senior leaders to triple-check the work.
- Get out of the IT integration business. For a financial services business for example, mastering the art of installing Linux, or building expertise in integrating business process management (BPM) with MQ provides no sustainable differentiation in the business; they needed to push the burden of delivering stable middleware to their vendors & partners, so that their employees can focus on higher-value technology problems within financial services.
- Even though they may source from the best vendors in the industry, companies still struggle to move at market speed, because the problem isn’t technology; rather it’s the processes that stifle them, and the fire drills that are distracting the top talent from focusing on higher-value problems. Orgs need a catalyst to break existing processes, so they can rebuild them to enable them to move faster.
Hybrid cloud can be a keystone of an application platform, responsible for creating, hosting, and managing application “patterns”. Patterns are pre-defined templates that when combined with just-in-time configuration, become complete & runnable applications. These runnable applications are composed of business logic as well the required application services – application servers, BPM, MQ, business rules, mobile services, web API gateways, etc. Working with vendors, the Enterprise Architecture team can build a library of runnable patterns that implement standards for high availability, disaster recovery, security, and logging. They are then able to define a pattern once, and guarantee that every application in the enterprise is making use of the standardized services.
The most important innovation in the solution is the library of application patterns built. The library of patterns should be integrated with the automation services delivered by the Integrated Systems-based hybrid cloud. The patterns are further optimized and resilient by leveraging elasticity, workload management, and health management services. This integration ensures that every instance of the pattern is running and available. If a node in the cloud fails, the health management services will automatically take action – moving the pattern to a working node, with zero impact to the application availability.
Each application pattern has 3 topologies:
- Development — which is usually a single virtual machine.
- Test — which is usually a highly available pair of clustered virtual machines.
- Production — a highly available pair of clustered virtual machines, integrated with disaster recovery processes.
The only difference between Test and Production patterns are the user id’s, passwords, and IP addresses; this is how organizations ensure consistency and repeatability as applications progress through the development process, avoiding situations where an application works in test, but doesn’t work in production.
In my next post I’ll discuss the remaining key capabilities to ensure success. Individually, the hybrid cloud, continuous delivery, and continuous insights capabilities will incrementally move the needle. By combining these three technologies in innovative ways, organizations can truly transform how they operate.
CTO, Splunk Inc.
Earning a Seat at the Table: Introduction
Earning a Seat at the Table: Responsibly Move at Market Speed
Earning a Seat at the Table: Hybrid Cloud with Continuous Delivery & Insights (Part 2)