orxfun Goto Github PK

followers: 1.0 following: 6.0 repos: 24.0 gists: 0.0

Name: Ugur Arikan

Type: User

Bio: Operations Research scientist / practitioner, optimization, mathematical programming, algorithms, systems design, etc. recently mostly in Rust.

Location: Germany

Blog: https://orxfun.github.io/cv/

Hi, I'm Uğur Arıkan

👋 𝙰𝚋𝚘𝚞𝚝 𝙼𝚎

Operations Research (OR) Scientist / Practitioner, studied in the Middle East Technical University, postdoc in Singapore University of Technology and Design (resume)
Currently located in Bonn, Germany, working at DHL
💬 Reach me [email protected] 〰️ https://www.linkedin.com/in/uarikan/
Projects available at https://github.com/orxfun mostly in rust & c#.

🤟 I like

all things OR	programming languages
optimization	⭐ rust, c#
algorithms & data structures	👍 react, ts, c++
multiobjective decision making	👌🏽 go, f#
speed & efficiency	🤔 zig

🎈 𝙸'𝚖 currently 𝚞𝚙 𝚝𝚘

An expressive, efficient and productive mathematical programming / modeling crate for rust. why?
- I appreciate how early good_lp allowed us to build mathematical models so that we could have the joy of using rust in OR projects.
- Currently, I am working on an alternative modeling library.
- A macro-free and concise api which does not require more lines than model-on-paper has.
- You may see a demo in C# below, and find design choices of the underlying library here.
- Targeting similar design choices with rust as summarized below:


concise	simple	solver agnostic	reusable & composable model components
immutable	type safe	separation of model from data	abstraction over inputs

Working to make self referential collections convenient & efficient while staying safe. why?
- Such collections are common building blocks of data structures used in many algorithms.
- They can be be implemented unsafely, in unsafe rust or c++. They can also be implemented safely in garbage collected languages. Can we get the best of both, i.e., safe and efficient rust implementation?
- Towards this target, I worked on orx-pinned-vec & orx-fixed-vec & orx-split-vec & orx-imp-vec & orx-selfref-col.
- This allowed me to build orx-linked-list with regular & references and without any unsafe pointer access, unlike the std implementation.
- Efficient & flexible trees 🌴 and graphs are in progress.
Recently, I started playing around with simple and efficient concurrent data structures, such as orx-concurrent-bag. The target is to use these as building blocks of an alternative lightweight (maybe no-std) and efficient parallelization library.
Also working on efficient data structures as I need in algorithms, such as orx-priority-queue.
And trying some experimental functional ideas such as orx-closure and orx-funvec.

mathematical modeling in action 🔎

Ugur Arikan's Projects

orx-bench-shortest-path

Benchmarks for single source single sink shortest path implementations

orx-closure

An explicit closure with absolute seperation of the captured data from the function.

orx-concurrent-bag

An efficient, convenient and lightweight grow-only concurrent data structure allowing high performance concurrent collection.

orx-concurrent-iter

A thread-safe, ergonomic and lightweight concurrent iterator trait and efficient implementations.

orx-concurrent-ordered-bag

An efficient, convenient and lightweight grow-only concurrent data structure allowing high performance and ordered concurrent collection.

orx-concurrent-vec

An efficient, convenient and lightweight grow-only read & write concurrent data structure allowing high performance concurrent collection.

orx-fixed-vec

An efficient fixed capacity vector with pinned elements.

orx-fun-compose

A function composition library for C#, heavily experimental right now, would be more practical if C# gets generic operator overloading.

orx-fun-funvec

Unified functional collections allowing access by index.

orx-fun-option

An option type for C# aiming to be explicit while concise.

orx-fun-result

A simple result type for C#, not minding the error type, instead aiming to be explicit, concise and fluent.

Traits to unify access to elements of n-dimensional vectors which are particularly useful in algorithms requiring both flexibility through abstraction over inputs and performance through monomorphization.

orx-imp-vec

A data structure to enable conveniently building performant self referential collections such as trees, graphs and linked lists.

orx-linked-list

An efficient and recursive singly and doubly linked list implementation.

orx-mathprog-gallery

Gallery of mathematical models by orx-mathprog of well-known problems.

orx-parallel

placeholder

orx-pinned-concurrent-col

A core data structure with a focus to enable high performance, possibly lock-free, concurrent collections using a PinnedVec as the underlying storage.

orx-pinned-vec

PinnedVec trait defines the interface for vectors which guarantee that elements added to the vector are pinned to their memory locations unless explicitly changed.

orx-priority-queue

Priority queue traits and efficient d-ary heap implementations.

orx-selfref-col

`SelfRefCol` is a core data structure to conveniently build safe and efficient self referential collections, such as linked lists and trees.

orx-split-vec

An efficient constant access time vector with dynamic capacity and pinned elements.

orx-tree

go green

orxfun Goto Github PK

Hi, I'm Uğur Arıkan

👋 𝙰𝚋𝚘𝚞𝚝 𝙼𝚎

🤟 I like

🎈 𝙸'𝚖 currently 𝚞𝚙 𝚝𝚘

Ugur Arikan's Projects

Recommend Projects

Recommend Topics

Recommend Org