> ## Documentation Index
> Fetch the complete documentation index at: https://docs.observability.platform.ai71services.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Key Engineering Considerations

Below are the key engineering considerations that led us evaluating best fit and moving towards to a self-hosted ELK stack.

<CardGroup cols={1}>
  <Card title="Enhanced Governance and Compliance" icon="users-between-lines">
    Gaining greater control over data security, privacy, and regulatory compliance.
    <li>Meeting regulatory requirements (e.g., local UAE regulatory compliance) is primary may demand that data is stored within specific geographical locations or infrastructure under your control.</li>
  </Card>

  <Card title="Data Ownership and Control" icon="database">
    Maintaining complete ownership of all observability data.
    <li>Should gives us ability to manage data, retention policies, and access and privacy. And implementation of end to end data life cycle.</li>
  </Card>

  <Card title="Meeting Customer Expectations" icon="user">
    Addressing customer concerns regarding data privacy and security.

    <li>Should be able to achieve/address customer expectations and their concerns in keeping their data secure and maintaining data privacy.</li>
  </Card>

  <Card title="Cost Optimization" icon="money-bill-wave">
    Shifting from an OPEX to a CAPEX model for observability infrastructure as the subscription based models costs high due to high with volume of logs,application data streams.
    <li> Unlike the subscription based models like datadog, A self-hosted ELK stack shifts the cost model from OPEX to CAPEX (capital expenditure) for infrastructure, offering potential long-term cost savings, especially with substantial data volumes. This decision implies an investment in internal expertise for managing the ELK stack.</li>
  </Card>

  <Card title="Avoiding Vendor/data Lock-in" icon="lock">
    Gaining flexibility in choosing observability tools and technologies.

    <br />

    <li>It allows us with the freedom to own and manage out own data life cycle, customize the stack, integrate with other open-source tools, and switch vendors if needed.</li>
  </Card>
</CardGroup>