In-Memory Data Grids (IMDGs) process complex data at high speeds, generally for large-scale implementations, with parallelized distributed processing. Utilizing a unified API, and collocating the applications and data in the same memory space, the IMDG minimizes latency and maximizes performance, distributing data and workloads across computers within a network.
GigaSpaces cloud-native IMDG utilizes Space-Based Architecture (SBA), in which the data grid topology is composed of Spaces clusters. Spaces are application clusters, each of which is comprised of a number of self-sufficient Processing Units (PUs). Each PU is responsible for processing the services and data that are sent to the space partition that it runs.
Co-locating data and applications within IMDGs enables enhanced scalability of both the data and applications, and reduces latency and bottlenecks.