Jumanji: The Case for Dynamic NUCA in the Datacenter

Schwedock, Brian C.; Beckmann, Nathan

doi:10.1109/micro50266.2020.00061

Cited by 7 publications

(1 citation statement)

References 77 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Jenga [91] extends Jigsaw to manage DRAM cache banks and to consider memory bandwidth in the algorithm. Jumanji [83] extends Jigsaw in the context of data centers, improving tail latency by reserving enough cache space for latency-critical applications to meet their deadlines, and reinforcing security by placing data from untrusted applications in different LLC banks. The main limitation of OS-based approaches is that they suffer from inaccurate classification of shared and private pages.…”

Section: B Software-managed Nuca Cachesmentioning

confidence: 99%

TD-NUCA: Runtime Driven Management of NUCA Caches in Task Dataflow Programming Models

Caheny

Alvarez

Casas

et al. 2022

SC22: International Conference for High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

In high performance processors, the design of onchip memory hierarchies is crucial for performance and energy efficiency. Current processors rely on large shared Non-Uniform Cache Architectures (NUCA) to improve performance and reduce data movement. Multiple solutions exploit information available at the microarchitecture level or in the operating system to optimize NUCA performance. However, existing methods have not taken advantage of the information captured by task dataflow programming models to guide the management of NUCA caches.In this paper we propose TD-NUCA, a hardware/software codesigned approach that leverages information present in the runtime system of task dataflow programming models to efficiently manage NUCA caches. TD-NUCA identifies the data access and reuse patterns of parallel applications in the runtime system and guides the operation of the NUCA caches in the hardware. As a result, TD-NUCA achieves a 1.18x average speedup over the baseline S-NUCA while requiring only 0.62x the data movement.

show abstract

Section: B Software-managed Nuca Cachesmentioning

confidence: 99%