SearcharxivSearch

arXiv subjects

Christos Kozanitis

Publications and source records attributed to Christos Kozanitis.

2 recordsLinked to original sources

NET4EXA: Pioneering the Future of Interconnects for Supercomputing and AI

NET4EXA aims to develop a next-generation high-performance interconnect for HPC and AI systems, addressing the increasing demands of large-scale infrastructures, such as those required for training Large Language Models. Building upon the proven BXI (Bull eXascale Interconnect) European technology used in TOP15 supercomputers, NET4EXA will deliver the new BXI release, BXIv3, a complete hardware and software interconnect solution, including switch and network interface components. The project will integrate a fully functional pilot system at TRL 8, ready for deployment into upcoming exascale and post-exascale systems from 2025 onward. Leveraging prior research from European initiatives like RED-SEA, the previous achievements of consortium partners and over 20 years of expertise from BULL, NET4EXA also lays the groundwork for the future generation of BXI, BXIv4, providing analysis and preliminary design. The project will use a hybrid development and co-design approach, combining commercial switch technology with custom IP and FPGA-based NICs. Performances of NET4EXA BXIv3 interconnect will be evaluated using a broad portfolio of benchmarks, scientific scalable applications, and AI workloads.

cs.NI

Garbage Collection or Serialization? Between a Rock and a Hard Place!

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 heap. Therefore, frameworks temporarily move long-lived objects outside the managed heap (off-heap) on a fast storage device. Unfortunately, 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 managed heap for processing. In this paper, we propose TeraHeap, a system that eliminates S/D overhead and expensive GC scans for a large portion of the objects in big data frameworks. TeraHeap relies on three concepts. (1) It eliminates S/D cost by extending the managed runtime (JVM) to use a second high-capacity heap (H2) over a fast storage device. (2) It reduces GC cost by fencing the garbage collector from scanning H2 objects. (3) It offers a simple hint-based interface, which allows frameworks to leverage knowledge about objects for populating H2. We implement TeraHeap in OpenJDK and evaluate it with 15 widely used applications in two real-world big data frameworks, Spark and Giraph. Our evaluation shows that for the same DRAM size, TeraHeap improves performance by up to 73% and 28% compared to native Spark and Giraph, respectively. Also, it provides better performance by consuming up to 8x and 1.2x less DRAM capacity than native Spark and Giraph, respectively. Finally, it outperforms Panthera, a garbage collector for hybrid memories, by up to 69%.

cs.PL