Code Optimization and Parallelization

Is your Python, R or C/C++ code too slow or does it require a lot of memory?

We can assist you in profiling your code, identifying the pain points, and assessing the feasibility of techniques to reduce the runtime and/or the memory footprint.

Example of techniques used to reduce the runtime:

• Memorization, i.e., storing the result of expensive function calls
• Reduction of the memory swapping
• Usage of better data structures (e.g., lists, sets, hash tables)
• Usage of performant libraries (e.g., OpenMP, Cython, Numba, NumPy, Pandas, dplyr)
• Parallel, distributed, and GPU computing (e.g., MPI, IPython Parallel, Dask) on the desired infrastructure, for example ScienceCloud or ScienceCluster
• Functional and domain decomposition

Examples of techniques used to reduce the memory footprint:

• Usage of better data structures (e.g., lists, sets, hash tables)
• Out of memory processing combined with specialized file formats (e.g., memory mapped files, HDF5, SQLite)
• Lazy evaluation and data batch processing
• Distributed computing (e.g., MPI, IPython Parallel, Dask)

• Get in contact with Science IT and explain briefly what type of service you are interested in together with a short description of your use case and needs. contact Science IT
• Based on the information you provided, your request will be routed to a suitable expert, who gets in contact with you to follow up on the remaining steps.

• Efforts-based costs
• Expert services are restricted to UZH researchers and groups.
• More details (incl. costs and agreement templates) are available in the UZH Intranet.