Dynamic reconfiguration of ASAP pipelines

Understand how and when to reconfigure Application Semantic-Aware Processing (ASAP) primitives and pipelines in production when workload information is unavailable or workload structure changes after deployment, potentially using ASAP itself to inform reconfiguration policies.

Background

The paper’s prototype pipelines are configured once at startup using a priori information about workload structure. The authors note that production deployments may lack this information or may experience workload changes after deployment. A fully realized ASAP framework therefore requires policies for deciding when reconfiguration is necessary, determining how primitives and pipeline stages should be reconfigured, and potentially exploiting ASAP-generated information to guide those decisions. The paper explicitly identifies this as an unresolved challenge.

References

Understanding how/when to reconfigure ASAP primitives and pipelines, and possibly using ASAP itself to inform these policies, is an open challenge.

— ASAP: Reimagining the Data Lifecycle using Application Semantic-Aware Processing  (2608.12735 - Srivastava et al., 13 Aug 2026) in Section 5, “Open Challenges,” subsection “Dynamic pipeline reconfiguration”

Processor-class mapping is static after compilation: \mycc selects the logical PU class, and \myrt only chooses among available instances of that class at runtime; adaptive remapping across PU classes is future work.

— Epic: Efficient Programming Paradigm for In-Storage Computing  (2609.19206 - Wang et al., 16 Sep 2026) in Section 3.7, “Scope and Limitations”