Scholarlink

<p>Memory tiering provides a&#xa0;promising approach to enhance the flexibility of system memory configurations in modern compute servers. Intel® Flat Memory Mode (FLAT) is a&#xa0;hardware-managed memory tiering system for CXL memory that operates at 64&#xa0;B cache-line granularity. In this paper, we evaluate the performance of FLAT using On-Line Analytical Processing (OLAP) workloads in SAP HANA, an in-memory database management system. Our results demonstrate that FLAT outperforms software-managed memory tiering and directly-attached CXL memory by leveraging cache-line granularity of data placement and exploiting data locality. For the first time, we evaluate FLAT with large-scale enterprise business warehouse workloads in SAP HANA, showing that it achieves performance comparable to a&#xa0;DRAM-only baseline, despite the high latency and limited bandwidth of CXL memory. Currently, FLAT accepts only a&#xa0;<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(1:1\)</EquationSource> </InlineEquation> ratio between local and CXL memory. When the local memory capacity exceeds that of CXL memory, Intel® Flat Memory Mode operates in a&#xa0;mixed mode (MIXED) containing flat memory and additional dedicated memory with remaining local memory in a&#xa0;separate NUMA node. Our evaluation of MIXED shows that performance degradation depends on access patterns and the degree of data locality exploited in flat memory. However, it requires software adjustments to utilize both dedicated memory and flat memory. To eliminate the need for such adjustments in MIXED, we propose a&#xa0;consolidated mode (CONSOLIDATED) that merges them into a&#xa0;single NUMA node. Our evaluation demonstrates that CONSOLIDATED results in negligible performance degradation, regardless of the ratio between dedicated memory and flat memory, and requires no software modifications at the application level.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Locality-Aware Memory Tiering with CXL for in-Memory Database Management Systems

  • 期刊论文

  • 发表日期：

    2026年6月23日

查看全文
  • Minseon Ahn,
  • Thomas Willhalm,
  • Donghun Lee,
  • Norman May,
  • Jungmin Kim,
  • Daniel Ritter,
  • Oliver Rebholz,
  • Yash Gupta,
  • Tarik Yuksek,
  • Byeonghun Hwang,
  • Kyumin Park,
  • Seungpyo Cho

摘要

Memory tiering provides a promising approach to enhance the flexibility of system memory configurations in modern compute servers. Intel® Flat Memory Mode (FLAT) is a hardware-managed memory tiering system for CXL memory that operates at 64 B cache-line granularity. In this paper, we evaluate the performance of FLAT using On-Line Analytical Processing (OLAP) workloads in SAP HANA, an in-memory database management system. Our results demonstrate that FLAT outperforms software-managed memory tiering and directly-attached CXL memory by leveraging cache-line granularity of data placement and exploiting data locality. For the first time, we evaluate FLAT with large-scale enterprise business warehouse workloads in SAP HANA, showing that it achieves performance comparable to a DRAM-only baseline, despite the high latency and limited bandwidth of CXL memory. Currently, FLAT accepts only a  \(1:1\) ratio between local and CXL memory. When the local memory capacity exceeds that of CXL memory, Intel® Flat Memory Mode operates in a mixed mode (MIXED) containing flat memory and additional dedicated memory with remaining local memory in a separate NUMA node. Our evaluation of MIXED shows that performance degradation depends on access patterns and the degree of data locality exploited in flat memory. However, it requires software adjustments to utilize both dedicated memory and flat memory. To eliminate the need for such adjustments in MIXED, we propose a consolidated mode (CONSOLIDATED) that merges them into a single NUMA node. Our evaluation demonstrates that CONSOLIDATED results in negligible performance degradation, regardless of the ratio between dedicated memory and flat memory, and requires no software modifications at the application level.