Multicloud is an architecture decision, not a trend

Running two clouds costs more than running one. It is worth it when it solves a concrete problem that a single cloud does not.

  • A service is better, or only exists, on another cloud

    For example, Oracle databases on OCI and an analytics or AI service on another cloud. Each workload goes where it performs best.

  • Continuity cannot depend on a single provider

    Critical systems with backup or recovery on a second cloud, tested and not just documented.

  • A regulation or a contract dictates where data lives

    Data residency requirements, from your industry or a client, that force a specific region or provider.

  • You already have more than one cloud

    Through acquisitions, decisions made by each area or inherited projects. What is missing is governing them as a single operation.

  • Your team cannot run two platforms

    Each cloud has its own identity, network, billing and ways of failing. Without the capacity to run them, multicloud adds risk.

  • The workload moves a lot of data between clouds

    Egress is billed and latency adds up. Sometimes the cheapest option is to keep the data and the applications that use it together.

  • The only goal is to avoid depending on one provider

    Without a concrete risk to mitigate, duplicating platforms duplicates costs. There are cheaper ways to reduce dependency, such as containers and infrastructure as code.

Three phases, the same team from start to finish

The three phases of our forward deployed engineering, designed for an operation spread across several clouds.

  1. Phase 1

    Discover and design

    We inventory your workloads, their dependencies and their costs, and decide with data which ones stay, which ones move and to which cloud.

  2. Phase 2

    Build and deliver

    We set up the shared foundation —identity, network, interconnection, security and infrastructure as code— and move workloads in stages, validating each one.

  3. Phase 3

    Operate and hand over

    We leave operations documented and measured across every cloud, and hand them over to your team or run them for you.

Independent of any cloud

We choose the cloud for your operation, not because of an agreement with a vendor. And because we are a forward deployed engineering firm, the same engineers who design the architecture build it and are accountable for it in production.

About forward deployed engineering

What people usually ask us

What is multicloud?

It means using services from more than one public cloud provider —for example, OCI and Azure— as part of the same operation. It does not mean duplicating everything on every cloud: it means placing each workload where it works best and governing them as a single system.

How is it different from hybrid cloud?

Hybrid cloud combines public cloud with your own infrastructure, such as a data center. Multicloud combines several public clouds. Many operations are both at once, and that is how we design them.

Is multicloud more expensive?

It can be if it is done without judgment: there are two platforms to run, and data transfer between clouds is billed. Done well, each workload goes where it costs less or performs better. That is why we run those numbers in the assessment, before recommending it.

How are the clouds connected to each other?

Over dedicated private links, such as FastConnect on OCI, Direct Connect on AWS and ExpressRoute on Azure, or through an interconnection provider such as Megaport, and with IPsec VPN when the volume allows it. Oracle and Microsoft also offer a direct interconnection between OCI and Azure in the regions where it is available.

Can you run our multicloud platform?

Yes. We offer managed operations in three tiers —Essential, Professional and Strategic— on your own accounts, or we hand operations over to your team with documentation and support.

Tell us which clouds you use today and what you want to solve

We reply within 24 business hours with a clear proposal and a defined scope.

Book an assessment