Propostas Atribuidas

DEI - FCTUC
Gerado a 2024-03-29 09:07:27 (Europe/Lisbon).
Voltar

Titulo Estágio

Monitoring Microservices Architectures

Áreas de especialidade

Engenharia de Software

Local do Estágio

DEI

Enquadramento

With increased reliance on the cloud and most
modern systems being designed as cloud native, and
given the trend in the industry to follow
service-oriented architectures, particularly
microservices, there is renewed interest in
monitoring the state of the system with minimal
interference to its operation.¬
¬
Current Application Performance Monitoring tools,
like New Relic, mostly collect infrastructure
measurements, like CPU occupation or latencies,
thus being able to pinpoint bottlenecks. However,
achieving a deeper insight of a large-scale
distributed architecture comprised of
microservices, requires other techniques, like
distributed tracing and snapshotting. These well
enable system administrators to drill down to the
level of the workflows of microservice invocations. However, distributed tracing and other monitoring
techniques might involve a heavy utilization of
instrumentation, with the corresponding needs for programming and configuration

Objetivo

In this internship, the student should focus on a
particular configuration of microservices, possibly
running in a containerized infrastructure, where
all microservices are served by an API Gateway
(similar to a reverse proxy, capable of providing
functionalities such as security and load
balancing), according to the following
illustration:¬
¬
Client --> API Gateway --> Microservice¬
¬
One should notice that the client will often be a
different microservice. The work of this internship
should explore a monitoring approach centered on
the API Gateway, using as little information from
the microservice as possible. ¬
¬
We intend to use the response times experienced by
clients (i.e., the invoking side), to infer about
the microservice performance. The goal is to detect
internal bottlenecks looking at the system from the
API Gateway in a nearly black-box monitoring. This
will involve among other challenges, the need to
configure machine-learning algorithms that run on
the API Gateway.

Plano de Trabalhos - Semestre 1

- Review of the state of the art and technologies on service registry and discovery
- Requirement analysis (including both functional and non-function requirements)
- System architecture
- Writing of the preliminary thesis

Plano de Trabalhos - Semestre 2

- System development
- System testing
- Writing of the final thesis

Condições

The student might receive a scholarship from the FCT DataScience4NP project (745€ / month).

Orientador

1) Filipe Araújo 2) Jorge Cardoso, 3) Rui Pedro Paiva,
filipius@dei.uc.pt 📩