一句话：提出 TeraHeap，缓解托管大数据框架在堆外缓存长生命周期对象时的高序列化开销与内存压力。
大数据分析框架（如 Spark 和 Giraph）需要处理并缓存海量数据，这些数据并不总能放入托管堆。因此，框架会将长生命周期对象临时移到快速存储设备上的堆外区域。然而，这一做法导致两个问题：一是高序列化/反序列化开销，二是当堆外对象被移回堆以进行处理时产生高内存压力。TeraHeap 针对上述问题，旨在降低托管大数据框架中的内存压力。给定摘要仅描述了问题与系统名称，未提供具体方法机制、实现细节或实验结果，因此无法进一步概括其设计、评估指标与结论。
Big data analytics frameworks, such as Spark and Giraph, need to process and cache massive amounts of data that do not always fit on the managed heap. Therefore, frameworks temporarily move long-lived objects outside the managed heap (off-heap) on a fast storage device. However, this practice results in (1) high serialization/deserialization (S/D) cost and (2) high memory pressure when off-heap objects are moved back to the heap for processing.
@inproceedings{paperbot10072,
title = {TeraHeap: Reducing Memory Pressure in Managed Big Data Frameworks},
author = {Iacovos G. Kolokasis and Giannos Evdorou and Shoaib Akram and Christos Kozanitis and Anastasios Papagiannis and Foivos S. Zakkak and Polyvios Pratikakis and Angelos Bilas},
booktitle = {ASPLOS},
year = {2023},
doi = {10.1145/3582016.3582045}
}