In Situ Pipeline

This is the newer, more general version of NEST in situ framework.

The in situ pipeline consists of a set of libraries that can be integrated into neuronal network simulators developed by the HBP to enable live visual analysis during the runtime of the simulation. The library called ‘nesci’ (neuronal simulator conduit interface) stores the raw simulation data into a common conduit format and the library called ‘contra’ (conduit transport) transports the serialized data from one endpoint to another using a variety of different (network) protocols. The pipeline currently works with NEST and Arbor. Support for TVB is currently in development.

Prototypical implementation into the HPAC Platform finalised in February 2019.

Date of releaseFirst released in July 2018 with continuous updates (see also above)
Version of software18.07
Version of documentation18.07
Software available
DocumentationSee the readme files in the repositories
ResponsibleRWTH: Simon Oehrl (
Requirements & dependenciesRequired: CMake, C++14, Conduit
Optional: Python, Boost, ZeroMQ
Target system(s)Desktops/HPC Systems running Linux, macOS or Windows


The development of T-Storm was co-funded by the HBP during the Ramp-up Phase. This page is kept for reference but will no longer be updated.

T-Storm is a platform for supporting scalable real-time analytics of massive sets of voluminous time-series. The platform is constructed over the Storm parallel dataflow engine, and supports both vertical scalability (fully utilizing high-end servers and multi-core systems) and horizontal scalability (scaling across a cluster of physical machines or even incorporating virtual cloud resources). The current version enables efficient maintenance of the highly correlated time-series in linear space and near-linear computational complexity (in practice, computational complexity depends on the input time-series). This functionality is, for example, useful to identify the pairs of neurons that fire in a correlated manner.

T-Storm is distributed as a prepared virtual machine. To use the platform, the user needs to 1) deploy and configure the required number of virtual machines, depending on the number of time series to monitor, and their velocity; 2) configure the virtual machines so that they have network access and can talk to each other; 3) provide the input in the documented format. Full instructions are provided with the virtual machine.

Date of releaseApril 2015
Version of software0.1
Version of documentation0.1
Software available
ResponsibleMinos Garofalakis (
Requirements & dependenciesJava (JDK>=7) (
Storm parallel dataflow engine (
Target system(s)Cluster