Search

Administered Applications on a Public Cloud: The Importance of Operations, Part II

Share it

Managing Deployed Software on a Shared Cloud: The Significance of Functions, Part Two

When delving into the realm of administering software on a communal cloud, one must consider the vital role operations play. In the previous installment, we scrutinized the essence of operational efficiency within public cloud setups. The key takeaway was the complexity and resource-intensive nature of achieving excellence in this domain. Given these challenges, you may be pondering ways to excel without draining all your resources on operations. Questions like “Can I still foster innovation?” or “Do I possess adequate resources to manage all this?” may be running through your mind. Fret not, as I am here to provide guidance through this intricate terrain.

Operational Oversight Alternatives

When it comes to managing your operational activities, you are faced with two primary choices: self-management or partnering with an external service provider. While the actual operations remain consistent regardless of your decision, certain options may be more favorable based on your current business stage and primary focus. Let’s delve into what each alternative entails for your organization.

Essential Criteria

As mentioned earlier, the operations themselves and the prerequisites for attaining operational excellence are akin irrespective of your chosen path. While it may be challenging to precisely outline all the necessities for operating your public cloud clusters at peak efficiency, there are a few fundamental requirements that are imperative for the smooth functioning of any public cloud deployment. These include:

Personnel

Individuals are the linchpin of any project, particularly in the context of public cloud deployments. However, in this scenario, the requirement isn’t for just any individual but for seasoned software engineers. It’s arduous to pinpoint the exact team size needed, but a general industry guideline suggests that one engineer should typically oversee the operation of fewer than 100 nodes or clusters.

Attracting and retaining talent can prove to be a daunting task. Considering the prevalent shortage of software engineering professionals in the industry, competition for skilled engineers is fierce. Engineers are typically driven by purpose, challenge, scope, and competitive remuneration, all of which are well-deserved incentives. Competent engineers can be the determining factor between success and failure for entire businesses. Simply put, operations are futile without the right people, and exceptional operations hinge on having the right talent.

Time

While automation plays a vital role in management, there are certain aspects of monitoring, alerting, and incident resolution that necessitate manual intervention. This translates to a significant portion of operations requiring substantial time investments from engineers. Ideally, senior team members should focus on innovation and addressing intricate issues, while junior and intermediate team members handle routine operational tasks. However, the availability of time is a scarce yet indispensable resource for achieving operational excellence, subject to fluctuations based on project demands.

Knowledge

Assuming you have a proficient team in place with ample time allocated for operational tasks, the realization may strike upon deployment – the dynamic and volatile nature of public cloud operations necessitates a continuous focus on nurturing and refining operational skills. Knowledge is the lifeblood of operations, akin to natural pearls in jewelry: rare, invaluable, demanding to acquire, and even harder to sustain. Fostering a culture of ongoing learning and skill enhancement within an engineering team is imperative. To ensure operational environments function optimally, yield dependable outcomes, and align with industry trends and demands, continuous learning and skill enhancement are non-negotiable. Knowledge should be a cornerstone value for any operational team, and arguably any team in general.

Financial Resources

Evidently, the preceding criteria are intuitive and logical. However, the emphasis lies in highlighting that all these aspects necessitate substantial financial investments from the company embarking on the formation and cultivation of an operational team. The costs associated with recruiting, retaining, training, and supporting talent are substantial. Unforeseen costs are prevalent in this context. Anticipating the headcount, training requirements, and time constraints with precise certainty is challenging. Therefore, astute financial planning is imperative to ensure seamless operational progress.

Adopting a DIY Approach

Now let’s explore the avenues available for allocating resources to operate a public cloud application deployment. Embarking on a do-it-yourself journey is commendable as it embodies a spirit of self-reliance. Assuming responsibility for every facet of the project implies steering its trajectory. This approach necessitates assembling a team of operational experts to manage your public cloud application environment from inception to decommissioning. As emphasized earlier, it’s crucial to nurture and upskill this team continuously to align with the dynamic public cloud ecosystem.

Success Imperatives

Achieving success through the DIY route is commendable. However, certain factors enhance the likelihood of excelling and truly prospering in this endeavor. Opting for a DIY strategy is particularly advisable if your organization is tech-centric, signifying a primary focus on software or hardware. A tech-centric orientation not only appeals to top engineering talent on ideological grounds but also facilitates leveraging the collective prowess of all teams to ensure technical robustness. Additionally, top-tier professionals gravitate towards intellectual challenges and autonomy for innovation, attributes that a tech-focused organization can provide.

Even if technology isn’t your core forte, venturing into self-managing your public cloud clusters either independently or in collaboration with your in-house IT team is feasible. Let’s assess the merits and demerits of the self-management approach.

Advantages

  • Complete Autonomy – Assume full control over decision-making processes, resource allocation, team size, training programs, and operational protocols.
  • Operational Liberty – Self-management affords the freedom to dictate monitoring processes, incident resolution methods, and compliance adherence.

Disadvantages

  • Unpredictable Expenses – Prepare for the unpredictability associated with hiring, training, and retaining an in-house operational team. The dynamic market landscape may entail unforeseen challenges such as rapid staff turnover.
  • Resource Procurement Hurdles – Sole responsibility for resource procurement entails ensuring engineers receive requisite training, tools and systems are up-to-date, and talent remains competent.
  • Accountability Challenges – The primary hurdle with self-management is assuming sole accountability for any lapses in operational efficacy. Limited external support in crisis scenarios underscores the importance of supplementary measures like opting for Enterprise Support.

Choosing a Managed Service Provider

Opting for the services of a managed service provider (MSP) is a straightforward proposition. Instead of internalizing operational responsibilities, you engage a reputable company to manage your environments according to your specifications at a fixed cost.

In my capacity as a product manager for managed services, I’ve encountered misconceptions surrounding the selection of managed services for open-source ecosystems. There’s an unfounded apprehension that relying on managed services reflects inadequacy in skill. However, the reality is quite the opposite. Embracing managed services liberates your engineering resources to focus on innovation, a tribute to their competencies and expertise. While operations are a foundational aspect supporting innovative projects, they hold minimal direct impact on innovation itself. Thus, entrusting talented engineers to drive innovation directly underscores corporate maturity and respect for their capabilities.

Nevertheless, managed services aren’t a one-size-fits-all solution. Similar to self-management, there are pros and cons:

Advantages

  • Resource Assurance – MSPs ensure a proficient team with suitable certifications and training manages your operations. With a primary focus on operational management, engineers devote the bulk of their time to honing their operational acumen.
  • Predictable Expenditure – While debating whether managed services are cost-effective compared to self-management is inconclusive, predictability stands out as a fundamental advantage. A transparent pricing model based on metrics assures fixed costs, alleviating concerns about unforeseen financial burdens.
  • Shared Responsibility – In the event of operational setbacks under MSP management, raising a ticket suffices for intervention. Reputable MSPs possess robust monitoring and alert mechanisms, promptly addressing issues before they escalate. Moreover, MSPs bear accountability for safeguarding your public cloud clusters, shielding your business from operational disruptions.

Disadvantages

  • Initial Financial Outlay – The sizeable pricing structures of MSPs can be daunting initially. However, smaller MSPs may cater to your operational needs without exorbitant costs. Anticipate an upfront operational management fee, scalable based on your deployment scope.
  • Limited Protocol Control – MSPs typically dictate compliance levels and incident resolution methodologies. While outlining specific requirements pre-contract is vital, beyond that, MSPs retain autonomy over operational protocols, which may entail adherence to global standards that occasionally constrain project flexibility.
  • Restricted Configuration Flexibility – MSPs assume full control over managed environments, with minimal client intervention allowed. Significant validation processes precede operational management, ensuring continuity throughout the contract term without scope for on-the-fly adjustments.

Key Considerations

Before opting for an MSP, it’s crucial to evaluate various aspects. Is the MSP sufficiently equipped to handle your environment? Do they possess requisite experience and flexibility? Does their service model align with your operational requirements? Addressing these queries and more is imperative before committing to an MSP relationship. For an in-depth exploration of these factors, particularly within the context of AI infrastructure management, refer to my whitepaper, “An Executive Guide to Managed AI Infrastructure.” Subsequent posts will further dissect these considerations.

Concluding Insights

Let’s address the elephant in the room – my inherent bias. My professional trajectory focuses on aiding enterprises in attaining operational excellence through managed services. Are managed services a fail-safe route to realizing your business objectives? Not entirely. Can they expedite your progress? Undoubtedly. While success sans managed services is plausible, it may demand more financial, temporal, and overall effort investments. This underscores my belief in the efficacy of operational management.

To explore the services Canonical offers within this domain, visit our website at: https://ubuntu.com/managed

Until our paths cross again, stay vigilant!

Adrian

🤞 Don’t miss these tips!

🤞 Don’t miss these tips!

Solverwp- WordPress Theme and Plugin