# FOSDEM 2023

## Track: Fast and Streaming Data devroom
https://fosdem.org/2023/schedule/track/fast_and_streaming_data/

### Session: Running Real-time Stream Processing Analytics On Traces

scale as bottleneck.
Produce alets or trends about the data.
real-time data landscape (slide). Hard to decide which tools to use for an application.:

* stream processing engines
* Hazelcast platform: opensource https://hazelcast.com/open-source-projects/

### Session: CDC Stream Processing with Apache Flink

building blocks of stream processsing
- streams
- time
- state
- snapshots

distinct because:
- parallelization
- optional state storage
- snapshopts of datapiline topologies

APIs
- datastream
- table
- built on dataflow runtime
- working with flink sql: use dynamic tables (those are considered streams)

### Session: Ingesting over a million rows per second on a single instance

- specialize on timeseries, nothing else
- does ok for goespatial data
- optimization of inseting data to DB, while not compromising reading data
- 1 million rows per second
- keywords for grouping by time
- spatial joins to join based on time (ASOF JOIN)
- bechmarks for timeseries: timescale/tsbs

### Session: Building A Real-Time Analytics Dashboard with Streamlit, Apache Pinot, and Apache Pulsar

- real-time analyics (an area of applications for my phd)
- interest of: analyist, management, users
- photo,
- properties of real-time analytics sysems: photo
- demo: sourcecode in github
-

## Track: Continuous Integration and Continuous Deployment
https://fosdem.org/2023/schedule/track/continuous_integration_and_continuous_deployment/

### Session: How To Automate Documentation Workflow For Developers

- Cohesion in documentation: use a voice chart
- Voice and tone: put on a style guide

## Track: HPC, Big Data and Data Science
https://fosdem.org/2023/schedule/track/hpc_big_data_and_data_science/

### Session: Simplifying the creation of Slurm client environments

- working with slurm + containers
- Straw: python based tool to manage autentication between slurm and external containers.
- https://github.com/pllopis/straw

### Session: Troika: Submit, monitor, and interrupt jobs on any HPC system with the same interface

- Project Destination Earth
- troika: python packages to handle connections to a remote system, prepare and submit jobs for multiple HPC facilities. Easier to port workflows to different HPC facilities: https://pypi.org/project/troika/
- https://fosdem.org/2023/schedule/event/troika_hpc_jobs/

### Session: Reproducibility and performance: why choose?
CPU tuning in GNU Guix

- https://hpc.guix.info/about/
- Mutiversion packaging for optimization of hpc hardware architecture

### Session: LIBRSB: Universal Sparse BLAS Library

- https://fosdem.org/2023/schedule/event/librsb/
- dealing with sparse matrices
- for iterable methods

### Session: numba-mpi

- makes possible to couple numba and MPI
- https://pypi.org/project/numba-mpi/
- https://fosdem.org/2023/schedule/event/numba_mpi/

## Track: Python
https://fosdem.org/2023/schedule/track/python/

### Session: How to build an event-driven application in Python

- Memphis.dev
- define events -> define actions
- https://github.com/yanivbh1/python-event-driven-app
- https://fosdem.org/2023/schedule/event/python_build_event_driven_application/

### Session: Continuous Documentation for Your Code

- How to guides, tutorials, : https://diataxis.fr/

## Other session of Interest
-https://fosdem.org/2023/schedule/event/python_build_event_driven_application/

## Tools of Interest
- https://fosdem.org/2023/schedule/event/openresearch_pimmi/
- https://github.com/questdb/questdb
- https://streamlit.io/
- check slides of Pulsar
- click-house local
- https://vale.sh/
- haystack python