Skip to content
back to the presentation
Talk summary2023

Fintech Trends: Cloud Technologies

In this fintech-school lecture, cloud services provide a way to divide responsibility for infrastructure and a financial product. Alexander Polomodov explains the layers of cloud services, connects them to platform teams, and then considers financial-market business models. Choosing between owned capacity and a provider depends on cost, resource availability, and system requirements.

Fintech School 20233 min read

This summary is shorter than the original material; examples and assessments reflect the author’s experience and views.

The main thread of the material
01

From hardware to a business process

A cloud provides a pool of resources that can be allocated and reconfigured as demand changes. Drawing on van Steen and Tanenbaum, the lecture distinguishes four layers: hardware, virtualized infrastructure, platforms, and applications. Application users do not manage server cooling, and developers using a platform service need not know which physical machine holds an object. The layers define different boundaries between customer and provider responsibilities.

IaaS, PaaS, and SaaS map onto these layers. FaaS moves execution of a function into the provider's environment. A Bank of Russia paper adds BaaS for the financial market. Its distinction from banking SaaS concerns what the customer buys: a ready-made accounting or operational process, rather than an application in which the customer still needs to establish that process.

02

What a platform gives a product team

A hierarchy-of-needs analogy explains the motivation: the business wants useful applications, while hardware, infrastructure, and an execution environment remain necessary underneath. A platform team takes responsibility for that foundation so that product teams can concentrate on business features. The benefit is not merely moving servers elsewhere; it is giving developers a usable interface to capabilities they would otherwise have to assemble themselves.

Kubernetes illustrates this approach through declarative deployment, an established ecosystem, and managed offerings in public clouds. Building and maintaining an internal platform is nevertheless complex. The lecture does not turn Kubernetes into a universal recommendation: a large organization must weigh the value of a shared foundation against the development and operating effort required to sustain it.

03

Financial-market roles and infrastructure constraints

Using the Bank of Russia's materials, the author distinguishes a niche fintech company, a distributor, a product factory, and a technology company. Their capabilities complement one another: they develop financial services, bring competing offerings to customers, or provide technical integration. The resulting cloud ecosystem extends beyond renting compute capacity to services such as security, data exchange, and access to payment systems.

Placing critical banking operations in a cloud requires attention to regulation and the risks of outsourcing functions. Other factors include capital versus operating expenditure, capacity in the required region, non-functional requirements, and migration costs. The supplementary part introduces edge computing and different forms of distributed ledgers. These broaden the overview without removing the need to choose infrastructure for the requirements of a particular system.

Takeaways

What to take away

  1. 01Cloud service models differ in how much responsibility the provider takes, from infrastructure to a complete business process.
  2. 02A platform reduces repeated infrastructure work for product teams but creates its own maintenance responsibilities.
  3. 03Financial organizations need to assess cloud costs alongside regulation, regional capacity, and migration effort.

Sources