commit | 0888631043ba57d29e682367712619815ececc9d | [log] [tgz] |
---|---|---|
author | Matthias Boehm <mboehm7@gmail.com> | Fri Oct 30 22:08:13 2020 +0100 |
committer | Matthias Boehm <mboehm7@gmail.com> | Fri Oct 30 22:08:45 2020 +0100 |
tree | 589296fc25b386e63bab0c0ff2e1abafdc022fb6 | |
parent | 428016c38fb55a3b6094334c7c876b71bf4bf3f7 [diff] |
[SYSTEMDS-2708] Performance buffer pool handling of frames w/ strings This patch improves the performance of buffer pool handling for frames with string columns. There are many calls to getInMemorySize() on the path through the buffer pool. In case of string columns, these calls can all values to determine the lengths of individual strings. We now reuse previously computed size estimates within the frame block. Furthermore, this patch also adds generalize the asynchronous file cleaning in the buffer pool to now also accept serialization tasks (such as frames with string columns). However, because asynchronous serialization provides weaker guarantees this second modification is not enabled by default. On an example mini-batch scenario (with preprocessing in the individual iterations), this patch improved performance from 91s (86 right indexing incl bufferpool release) to 9.2s (5.1s right indexing). With the additional sync serialization, it further reduced to 7.9s (3.2s right indexing).
Overview: SystemDS is a versatile system for the end-to-end data science lifecycle from data integration, cleaning, and feature engineering, over efficient, local and distributed ML model training, to deployment and serving. To this end, we aim to provide a stack of declarative languages with R-like syntax for (1) the different tasks of the data-science lifecycle, and (2) users with different expertise. These high-level scripts are compiled into hybrid execution plans of local, in-memory CPU and GPU operations, as well as distributed operations on Apache Spark. In contrast to existing systems - that either provide homogeneous tensors or 2D Datasets - and in order to serve the entire data science lifecycle, the underlying data model are DataTensors, i.e., tensors (multi-dimensional arrays) whose first dimension may have a heterogeneous and nested schema.
Quick Start Install, Quick Start and Hello World
Documentation: SystemDS Documentation
Python Documentation Python SystemDS Documentation
Issue Tracker Jira Dashboard
Status and Build: SystemDS is renamed from SystemML which is an Apache Top Level Project. To build from source visit SystemDS Install from source