BAJU Meetup #8: Convex.jl and Distributed-Memory

1. Presenters: Karanveer Mohan and David Zeng

Topic: Convex.jl
Description: Convex.jl is a convex optimization modeling framework in Julia.Convex.jl translates problems from a user-friendly functional language into an abstract syntax tree describing the problem. This concise representation of the global structure of the problem allows Convex.jl to infer whether the problem complies with the rules of disciplined convex programming (DCP),and to pass the problem to a suitable solver. These operations are carried out in Julia using multiple dispatch, which dramatically reduces the timerequired to verify DCP compliance and to parse a problem into conic form. Convex.jl then automatically chooses an appropriate backend solver to solve the conic form problem.

2. Presenters: Jack Poulson

Topic: A proposal for distributed-memory direct linear algebra in Julia
Description:

Despite the fact that distributed-memory computing has been popular since the early 1990's, only three open source libraries for distributed-memory analogues of LAPACK (Dongarra, Demmel et al.) ever emerged: ScaLAPACK (Dongarra, Demmel et al.), PLAPACK (van de Geijn et al.), and Elemental (P. et al.). The former started in the late 1990's, while the latter is a modern spiritual successor of PLAPACK which makes use of an interconnected family of simple element-wise data distribution schemes. This talk will provide a brief overview of the functionality of Elemental (the DistMatrix class and distributed-memory SVD, EVD, sparse-direct Cholesky, etc.) as well as ongoing efforts to provide idiomatic interfaces to several languages (e.g., C++11, C, Python, and, soon, Julia).