STREAMSTORE is a research project that improves the state of the art in elastic, scalable, and fault-tolerant stateful stream-processing systems (SPSs) that access large amounts of internal and/or external state, through the evolution of the elasticity, predictability, and availability of scalable key-value store (KVS) technology beyond the state of the art. STREAMSTORE's targeted scientific breakthrough is the achievement of a coordinated SPS-KVS platform with aligned data-placement and task-placement policies across SPS and KVS for high performance and rapid recovery, leveraging a highly elastic scalable KVS optimized for RDMA networking.




STREAMSTORE is supported by the Hellenic Foundation for Research and Innovation (HFRI) through a faculty grant (Grant ID: HFRI-FM17-1998).