We are happy to announce the release of the latest version (2.7.0) of our OMPi OpenMP compiler, with significant improvements and new features, including:
- A new module (cuda) that allows offloading to CUDA GPUs with compute capability
OMPi on the Jetson Nano
We are pleased to announce that OMPi now supports the NVIDIA Jetson Nano 2GB boards. OMPi is the first compiler to provide full OpenMP support including offloading to the board’s GPU.
Support for Jetson Nanos is included in the … [ read more ]
Announcing OMPi v2.5.0
We are happy to announce the release of the latest version (2.5.0) of our OMPi OpenMP compiler, which includes a large number of improvements and new features such as:
- Task dependencies
- Doacross loops
- Support for all OpenMP 4.5 target-related
Parallax in effect
No parallax effect here. It is simply our new parallax machinery on top of its sibling, parade.
Parallax is a Dell EMC PowerEdge R740 server equipped with two Intel Xeon Gold 6130 CPUs, 64 GiB of main memory and … [ read more ]
We got a new toy!
Today we received parade, our new toy.
It is a cute Dell EMC PowerEdge R840 server stuffed with 64 cores (4 x 16-core Intel Xeon Gold 6130) and 256 GiB of main memory. It sure does not feel fluffy though!… [ read more ]
Diploma theses of Aggelos and Christos
On the previous two Thursdays (Oct. 17 and Oct. 24, 2019), Aggelos Kosintzis and Christos Karampeazis-Papadakis presented successfully their Diploma Theses. Aggelos surveyed benchmarking tools and suites related to OpenMP while Chris worked on HOMPI, our hybrid MPI/OpenMP task-based framework.… [ read more ]
Prestigious distinction for PPG
Our paper entitled “Adaptive Probabilistic Search in Complex Dynamic Networks” which was presented by Spiridoula Margariti in the 10th International Conference on Emerging Networks and Systems Intelligence (EMERGING 2018), has received the Best Paper Award!
The paper can be … [ read more ]