Deliverables

Work Package 2: Programming high-performance communication networks and fabrics in heterogeneous systems

Deliverable NumberDeliverable NameStateDate
D2.1 Report on Current Landscape for Novel Network Hardware
and Programming Models
Accepted31 January 2019
D2.2 Initial design document of MPI and GPI extensions for
heterogeneous systems with distributed GPUs and FPGAs
Accepted 30 June 2019
D2.3Final design specification and prototype implementation report of MPI
and GPI extensions for heterogeneous systems with distributed GPUs and FPGAs
Accepted 29 February 2020

Work Package 3: Efficient and Simplified Usage of Diverse Memories

Deliverable NumberDeliverable NameStateDate
D3.1 Report on Current and Emerging Transport Technologies Accepted31 January 2019
D3.2 Initial Design of a Memory Abstraction Device for Diverse Memories
Accepted31 May 2019
D3.3Final design specification and prototype implementation report of APIs
and runtime system for data placement, migration and access on diverse memories.
Accepted29 February 2020

Work Package 4: Productive computing with FPGAs, GPUs and low-power microprocessors

Deliverable NumberDeliverable NameStateDate
D4.1 Report on state of the art of novel compute elements
and gap analysis in MPI and GASPI
Accepted31 January 2019
D4.2 Design document of GPI-space extension for
distributed FPGAs and DSL for deep-learning applications.
Accepted31 May 2019
D4.3Final design specification and prototype implementation report of
GPI-Space extension for distributed FPGAS and DSL for deep-learning applications
Accepted29 February 2020

Work Package 5: Pilot applications and integration

Deliverable NumberDeliverable NameStateDate
D5.1 Report on Application Requirements and Roadmap Accepted31 January 2019
D5.2 Report on initial porting of applications to large-scale heterogeneous systems.
Accepted31 May 2019
D5.3Best practices document for porting applications to large-scale heterogeneous systemsAccepted29 February 2020

Work Package 6: Dissemination and Standardization