Data Science & Analytics

BIDMach - UC Berkeley

Description:

Fastest Big Data tools on the web. Holds the records for many common machine learning problems, on single nodes or clusters. Interactive environment for easily building and deploying machine learning models.

Features:

Scala interface. Supports linear regression, logistic regression, SVM, LDA, K-Means and other operations.

Multi GPU support:

Yes

Blazegraph GPU

Description:

First and fastest GPU accelerated platform for graph query. It provides drop-in acceleration for existing RDF/Sparql and Tinkerpop/ Blueprints graph applications. It provides high-level graph database APIs with transparent GPU acceleration for graph query.

Features:

GPU-accelerated SPARQL graph query, Data Management using the RDF interchange model, Tinkerpop/Blueprints Graph Support, Billions of edges on a single multi-GPU node, SaaS and Appliance models available.

Multi GPU support:

Yes

Blazegraph DASL

Description:

It marries the power and speed of CUDA. It delivers graph analytics at over 32 billion traversed edges per second and easily integrates with Spark and other data management platforms.

Features:

Scala-based graph analytic and machine learning application language, Ease of integration into Spark and Hadoop data ecosystems, Support for GPU cluster deployment.

Multi GPU support:

Yes

GPUdb

Description:

A distributed database for many core devices. GPUdb is a scalable, distributed database with SQL-style query capability for Big Data. Full suite of geospatial calculation capability.

Features:

Query against Big Data in real time. No pre-indexing allows for complex, ad-hoc query chains. Interactively explore large, streaming data sets.

Multi GPU support:

Yes

* Gunrock

Description:

Gunrock is a library for graph processing on the GPU. Gunrock achieves a balance between performance and expressiveness by coupling high performance GPU implementations with a high-level programming model, that requires minimal GPU programming knowledge.

Features:

Direction-optimizing BFS, SSSP, PageRank, Connected Components, Betweenness-centrality

Multi GPU support:

Yes

Jedox

Description:

Helps with portfolio analysis, management consolidation, liquidity controlling, cash flow statements, profit center accounting, treasury management, customer value analysis and many more applications, all accessible in a powerful web and mobile application or Excel environment.

Features:

This database holds all relevant data in GPU memory and is thus an ideal application to utilize the Tesla K40’s 12 GB on-board RAM. Scale that up with multiple GPUs and keep close to 100 GB of compressed data in GPU memory on a single server system for fast analysis, reporting and planning.

Multi GPU support:

Yes

MapD

Description:

MapD is GPU-powered big data analytics and visualization platform that is hundreds of times faster than CPU in-memory systems.

Features:

MapD uses GPUs to execute SQL queries on multi-billion row datasets and optionally render the results, all in milliseconds.

Multi GPU support:

Yes

* Sqream DB

Description:

GPU accelerated SQL database engine for big data analytics. Sqream speeds SQL analytics by 100X by translating SQL queries into highly parallel algorithms run on the GPU.

Features:

Up to 100TB of raw data can be stored and queried in a standard 2U server. Inserts and analyzes hundreds of billions of records in seconds. No indexes required. No changes to SQL code or data science paradigms required.

Multi GPU support:

Yes