Using remote memory for the Java heap enables big data analytics frameworks to process large datasets. However, the Java Virtual Machine (JVM) runtime struggles to maintain low network traffic during garbage collection (GC) and to reclaim space efficiently. To reduce GC overhead in big data analytics, systems group long-lived objects into regions and exclude them from frequent GC scans, regardless of whether the heap resides in local or remote memory. Recent work uses dual-heap design, placing short-lived objects in a local heap and long-lived objects in a remote region-based heap, limiting GC activity to local memory. However, these systems still scan the full remote heap to obtain liveness information, causing high network overhead.

In this paper, we propose SmartSweep, a system that uses approximate liveness information to balance network traffic and space reclamation in remote heaps. SmartSweep adopts a dual-heap design and avoids scanning or compacting objects in the remote heap. Instead, it estimates the amount of garbage in each region without accessing the remote heap and selectively transfers regions with many garbage objects back to the local heap for reclamation. Our preliminary results with Spark and Neo4j-GDS show that SmartSweep achieves comparable network traffic with TeraHeap, which lazily reclaims objects in the remote heap while reducing by up to 49% H2 peak memory usage, avoiding out-of-memory errors.