Reading club

Compiler Reading Club

Description

The ARCANA group organizes a weekly reading club where we discuss research papers.

Who: the compiler reading club is open to everyone

Where: the reading club is done in person and in the ARCANA lab (Mudd 3304/3308)

When: every Tuesday from 12:30pm to 2:00pm CST

What: we discuss

Compiler research papers

System research papers that do (or could) have a strong influence to compiler research

Protocol to choose the paper to review for a given week

You can propose a paper to review only the week before.
Deadline = Thursday 5pm (it's in the ARCANA calendar)

To propose a paper: send the link of the pdf of the paper to the #reading_club channel of the ARCANA slack group

You can vote to all papers you would like to review the week paper.
Deadline = Friday 5pm (it's in the ARCANA calendar)

To vote a paper: click Thumb up

on the slack message with the pdf link.

Friday at 5pm the paper with the highest votes will be read/discussed the week after.

Papers read

3/26/2024	Future Directions for Optimizing Compilers
2/20/2024	Latent Idiom Recognition for a Minimalist Functional Array Language using Equality Saturation
2/13/2024	egg: Fast and extensible equality saturation
1/30/2024	Sympiler: Transforming Sparse Matrix Codes by Decoupling Symbolic Analysis
1/16/2024	Propeller: A Profile Guided, Relinking Optimizer for Warehouse-Scale Applications
11/14/2023	Draft of one of our unpublished paper
10/31/2023	Generating Thread-Level Parallelism in Nondeterministic Programs
10/24/2023	Impact of Economics on Compiler Optimization
10/10/2023	Data Representation Synthesis
10/3/2023	Reducers and other Cilk++ hyperobjects
9/26/2023	A Transformation Framework for Optimizing Task-Parallel Programs
8/22/2023	Draft of one of our unpublished paper
8/15/2023	Draft of one of our unpublished paper
8/8/2023	Draft of one of our unpublished paper
8/1/2023	The tensor algebra compiler
7/18/2023	Indexed Streams: A formal intermediate representation for fused contraction programs
7/11/2023	Context Sensitivity without Contexts: A Cut-Shortcut Approach to Fast and Precise Pointer Analysis
6/27/2023	Mosaic: An Interoperable Compiler for Tensor Algebra
4/18/2023	HPVM: Heterogeneous Parallel Virtual Machine
3/21/2023	Looplets: A Language for Structured Coiteration
3/14/2023	Legion: Expressing Locality and Independence with Logical Regions
3/7/2023	Bridging Control-Centric and Data-Centric Optimization
2/21/2023	MPI-aware compiler optimizations for improving communication-computation overlap
2/14/2023	Scalable Automatic Differentiation of Multiple Parallel Paradigms through Compiler Augmentation
2/7/2023	HALO: Post-Link Heap-Layout Optimisation
1/24/2023	Automatic pool allocation: improving performance by controlling data structure layout in the heap
1/10/2023	Phloem: Automatic Acceleration of Irregular Applications with Fine-Grain Pipeline Parallelism
1/3/2023	Mesh: compacting memory management for C/C++ applications
12/13/2022	Beyond Induction Variables
12/6/2022	Accelerator-Level Parallelism
11/15/2022	Draft of one of our unpublished paper
11/1/2022	Draft of one of our unpublished paper
10/11/2022	The program dependence graph and its use in optimization
8/30/2022	Unified Analysis of Array and Object References
7/19/2022	Array SSA form and its use in parallelization
7/5/2022	An Efficient Profile-Analysis Framework for Data-Layout Optimizations
6/28/2022	Draft of one of our unpublished paper
6/21/2022	A Unifying Abstraction for Data Structure Splicing
6/14/2022	Dynamic dependency analysis of ordinary programs
6/7/2022	Memory SSA - A Unified Approach for Sparsely Representing Memory Operations
5/24/2022	Speculative Separation for Privatization and Reductions
5/10/2022	Unleashing the Hidden Power of Compiler Optimization on Binary Code Difference: An Empirical Study
3/29/2022	Halide: a language and compiler for optimizing parallelism, locality, and recomputation in image processing pipelines
2/22/2022	VeGen: a vectorizer generator for SIMD and beyond
1/25/2022	Region-Based Shape Analysis with Tracked Locations
1/18/2022	Progressive Raising in Multi-level IR
1/11/2022	OpenARC: open accelerator research compiler for directive-based, efficient heterogeneous computing
10/19/2021	Optimizing Indirect Memory References with milk
10/5/2021	Shape Analysis
9/28/2021	AddressSanitizer: a fast address sanity checker

Statistics

So far we have discussed 50 papers in the compiler reading club. Each paper is discussed for up to two hours.

Next is the number of papers discussed in the compiler reading club per year.

Year	Number of papers
2024	5
2023	23
2022	19
2021	3

Next is the number of papers discussed in the compiler reading club per month (accumulated over all years). This data is sorted to show the months that we read more (the months listed first on the table).

Year	Number of papers
January	8 (16.00%)
October	7 (14.00%)
February	6 (12.00%)
March	5 (10.00%)
June	5 (10.00%)
August	5 (10.00%)
July	4 (8.00%)
November	3 (6.00%)
May	2 (4.00%)
September	2 (4.00%)
December	2 (4.00%)
April	1 (2.00%)