Abstract: Scheduling techniques transform dataflow graphs (DFGs), for example, of digital signal processing (DSP) arrangements of filters, into efficient schedules for concurrent execution on processing resources coupled to a memory. A DSP arrangement may be represented by an executable model having interconnected filters represented by model elements. The techniques may apply scheduling transforms according to a classification of the model elements based on a lifetime of their internal states (e.g., finite or infinite). Exemplary scheduling transforms may include unfolding, coordinated loop scheduling and pipelining to parallelize a DFG and enhance overall performance, i.e., reduce average sample execution time of the DSP arrangement. Notably, the scheduling transforms may aggregate (i.e., merge) multiple finite state model elements for concurrent execution and repeat execution of infinite state model elements to achieve the overall improved performance.

Abstract: Techniques are described for tracking data objects transferred among multiple services in a computing environment. Services that are involved in the transfer of data objects may be instrumented to generate recordings that describe outbound and inbound transfers of data objects. The recordings may be analyzed to identify key-value pairs included in the transferred data objects, where the key corresponds to a data attribute that stores a particular value in a data object. For each pair of keys that are associated with a same or substantially similar value, a correlation metric may be updated for each instance of association. Over time, the correlation metric may indicate a higher degree of correlation for those pairs of keys that frequently share a same value. Pairs of keys exhibiting an above-threshold correlation count may be designated as related in that they are involved in the transfer of data objects.