CXL-Bench: Benchmarking Shared CXL Memory Access
- Marcel Weisgut,
- Daniel Ritter,
- Florian Schmeller,
- ,
- Tilmann Rabl
- Hasso Plattner Institute for Software Systems Engineering,
- SAP Research,
- ,
- ,
- ,
Publikation:
Konference artikel i Proceeding eller bog/rapport kapitel
Konferencebidrag i proceedings
Forskning
Peer-reviewOpen Access
Resume
Memory access paths between a CPU core and memory are increasingly complex. Data can be placed on local- or remote-socket memory, and on local- and remote-die memory on modern multi-die CPUs, affecting memory access performance. Cache-coherent inter-device interconnects, such as Compute Express Link (CXL), allow a CPU core to perform load and store instructions to memory of a peripheral device. Such accesses incur higher access latency than accesses to local-socket memory and increase the access path complexity. For database system developers, it is important to understand the performance implications of these complex memory architectures. In this work, we present CXL-Bench, a benchmark framework for quantifying access performance for different memory access paths. CXL-Bench provides many configuration options, such as memory access patterns, the operating system’s memory abstraction, cache bypass options, and a distributed mode for setups with multiple servers accessing memory of the same device. We demonstrate the utility of CXL-Bench by quantifying memory access characteristics of two servers accessing a shared CXL 1.1 memory device. Our results show that memory accesses of one server to the device affect the access performance of another server accessing the same device. On the other hand, memory (de)allocations using CXL memory configured as a character device complete quickly, making frequent re-allocation of CXL memory feasible.
Publikation information
Produktionstype
Publikation:
Konference artikel i Proceeding eller bog/rapport kapitel
Konferencebidrag i proceedings
Forskning
Peer-reviewOriginalsprog
EngelskPublikationsmilepæle
- Udgivet - 2025
Publikationsstatus
Udgivet - 2025
Titel på værtspublikation
International Workshop on Accelerating Analytics and Data Management Systems Using Modern Processor and Storage ArchitecturesFinansiering
We thank Seagate Technology LLC for their support and the anonymous reviewers for their feedback. This work was partially funded by SAP, the German Research Foundation (ref. 414984028), the European Union’s Horizon 2020 research and innovation programme(ref. 957407), and the Independent Research Fund Denmark’s Inge Lehmann program (grant agreement number 0171-00062B).
Adgang til dokumenter
Relateret event
Titel
Accelerating Analytics and Data Management Systems Using Modern Processor and Storage Architectures
Begivenhedstype
WorkshopGrad af anerkendelse
International begivenhedDato
01/09/2025 - 01/12/2025Lokation
LondonStorbritannien
