Polyvios pratikakis

SmartSweep: Efficient Space Reclamation in Tiered Managed Heaps

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 …

MaTSa: Race Detection in Java

Parallel programs are prone to data races, which are concurrency bugs that are difficult to track and reproduce. Various attempts have been made to create or incorporate tools that …

TeraHeap: Exploiting Flash Storage for Mitigating DRAM Pressure in Managed Big Data Frameworks

Big data analytics frameworks, such as Spark and Giraph, need to process and cache massive datasets that do not always fit on the managed heap. Therefore, frameworks temporarily …

TeraHeap: Reducing Memory Pressure in Managed Big Data Frameworks

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 …

Freeing Compute Caches from Serialization and Garbage Collection in Managed Big Data Analytics

Managed analytics frameworks (e.g., Spark) cache intermediate results in memory (on-heap) or storage devices (off-heap) to avoid costly recomputations, especially in graph …

Say Goodbye to Off-heap Caches! On-heap Caches Using Memory-Mapped I/O

DiSquawk: 512 cores, 512 memories, 1 JVM

Hierarchical Parallel Dynamic Dependence Analysis for Recursively Task-Parallel Programs

This work presents a hierarchical, parallel, dynamic dependence analysis for inferring runtime dependencies between recursively parallel tasks in the OmpSs programming model. To …

DiSquawk: 512 Cores, 512 Memories, 1 JVM

Building a Java™; Virtual Machine for Non-Cache-Coherent Many-core Architectures