HSCS: a hybrid shared cache scheduling scheme for multiprogrammed workloads

Abstract

The traditional dynamic random-access memory (DRAM) storage medium can be integrated on chips via modern emerging 3D-stacking technology to architect a DRAM shared cache in multicore systems. Compared with static random-access memory (SRAM), DRAM is larger but slower. In the existing research, a lot of work has been devoted to improving the workload performance using SRAM and stacked DRAM together in shared cache systems, ranging from SRAM structure improvement to optimizing cache tags and data access. However, little attention has been paid to designing a shared cache scheduling scheme for multiprogrammed workloads with different memory footprints in multicore systems. Motivated by this, we propose a hybrid shared cache scheduling scheme that allows a multicore system to utilize SRAM and 3D-stacked DRAM efficiently, thus achieving better workload performance. This scheduling scheme employs (1) a cache monitor, which is used to collect cache statistics; (2) a cache evaluator, which is used to evaluate the cache information during the process of programs being executed; and (3) a cache switcher, which is used to self-adaptively choose SRAM or DRAM shared cache modules. A cache data migration policy is naturally developed to guarantee that the scheduling scheme works correctly. Extensive experiments are conducted to evaluate the workload performance of our proposed scheme. The experimental results showed that our method can improve the multiprogrammed workload performance by up to 25% compared with state-of-the-art methods (including conventional and DRAM cache systems).

Keywords

multicore system / shared cache / workload performance

References

[1] Chou C, Jaleel A, Qureshi M K. BEAR: techniques for mitigating bandwidth bloat in gigascale DRAM caches. In: Proceedings of the 42nd Annual International Symposium on Computer Architecture. 2016, 198–210

[2] Lee Y, Kim J, Jang H, Yang H, Kim J, Jeong J, Lee J W. A fully associative, tagless DRAM cache. In: Proceedings of ACM/IEEE International Symposium on Computer Architecture. 2015, 211–222

[3] Hameed F, Bauer L, Henkel J. Adaptive cache management for a combined SRAMand DRAM cache hierarchy for multi-cores. In: Proceedings of the Conference on Design, Automation and Test in Europe. 2013, 77–82

[4] Hundal R, Oklobdzija V G. Determination of optimal sizes for a first and second level SRAM-DRAM on-chip cache combination. In: Proceedings of IEEE International Conference on Computer Design: VLSI in Computers and Processors. 1994, 60–64

[5] Qureshi M K, Loh G H. Fundamental latency trade-off in architecting DRAM caches: outperforming impractical SRAM-tags with a simple and practical design. In: Proceedings of IEEE/ACMInternational Symposium on Microarchitecture. 2012, 235–246

[6] Huang C C, Nagarajan V. ATCache: reducing DRAMcache latency via a small SRAM tag cache. In: Proceedings of International Conference on Parallel Architectures and Compilation. 2014, 51–60

[7] Hameed F, Bauer L, Henkel J. Reducing latency in an SRAM/DRAMcache hierarchy via a novel Tag-Cache architecture. In: Proceedings of the 51st Annual Design Automation Conference. 2014, 1–6

[8] Andrade D, Fraguela B B, Doallo R. Accurate prediction of the behavior of multithreaded applications in shared caches. Parallel Computing, 2013, 39(1): 36–57

[9] Manikantan R, Rajan K, Govindarajan R. Probabilistic shared cache management (PriSM). In: Proceedings of the 39th IEEE International Symposium on Computer Architecture. 2012, 428–439

[10] Wei W, Jiang D, Xiong J, Chen M. HAP: hybrid-memory-aware partition in shared last-level cache. In: Proceedings of IEEE International Conference on Computer Design. 2014, 28–35

[11] Holey A, Mekkat V, Yew P C, Zhai A. Performance-energy considerations for shared cache management in a heterogeneous multicore processor. ACM Transactions on Architecture & Code Optimization, 2015, 12(1): 1–29

[12] El-Moursy A, Sibai F N. V-Set cache: an efficient adaptive shared cache for multi-core processors. Journal of Circuits System & Computers, 2014, 23(23): 815–822

[13] Zhang D, Ju L, Zhao M, Gao X, Jia Z. Write-back aware shared lastlevel cache management for hybrid main memory. In: Proceedings of the 53rd Design Automation Conference. 2016

[14] Elhelw A S, Moursy A E, Fahmy H A H. Time-based least memory intensive scheduling. In: Proceedings of the 8th IEEE International Symposium on Embedded Multicore/Manycore Systems-on-Chip. 2014, 311–318

[15] Elhelw A S, El-Moursy A, Fahmy H A H. Adaptive time-based least memory intensive scheduling. In: Proceedings of the 9th IEEE International Symposium on Embedded Multicore/Manycore Systems-on-Chip. 2015, 167–174

[16] Chen Q, Zheng L, Guo M. DWS: demand-aware work-stealing in multi-programmed multi-core architectures. In: Proceedings of Programming Models and Applications on Multicores and Manycores. 2014

[17] Chen X, Xu C, Dick R P, Mao Z M. Performance and power modeling in a multi-programmed multi-core environment. In: Proceedings of the 47th Design Automation Conference. 2010, 813–818

[18] Roscoe B, Herlev M, Liu C. Auto-tuning multi-programmed workload on the SCC. In: Proceedings of International Green Computing Conference. 2013, 1–5

[19] Huang C, Ravi S, Raghunathan A, Jha N K. Synthesis of heterogeneous distributed architectures for memory-intensive applications. In: Proceedings of International Conference on Computer Aided Design. 2003, 46–53

[20] Huang C, Ravi S, Raghunathan A, Jha N K. Generation of heterogeneous distributed architectures for memory-intensive applications through high-level synthesis. IEEE Transactions on Very Large Scale Integration Systems, 2007, 15(11): 1191–1204

[21] Castellana V G, Ferrandi F. Abstract: speeding-up memory intensive applications through adaptive hardware accelerators. In: Proceedings of SC Companion: High Performance Computing, Networking Storage and Analysis. 2012, 1415–1416

[22] Yi W, Tang Y, Wang G, Fang X. A case study of SWIM: optimization of memory intensive application on GPGPU. In: Proceedings of International Symposium on Parallel Architectures, Algorithms and Programming. 2010, 123–129

[23] Athanasaki E, Anastopoulos N, Kourtis K, Koziris N. Exploring the performance limits of simultaneous multithreading for memory intensive applications. Journal of Supercomputing, 2008, 44(1): 64–97

[24] Chun K C, Jain P, Kim C H. Logic-compatible embedded DRAM design for memory intensive low power systems. In: Proceedings of IEEE International Symposium on Circuits and Systems. 2010, 277–280

[25] Jaleel A, Nuzman J, Moga A, Steely S C, Emer J. High performing cache hierarchies for server workloads: relaxing inclusion to capture the latency benefits of exclusive caches. In: Proceedings of IEEE International Symposium on High Performance Computer Architecture. 2015, 343–353

[26] Akin B, Franchetti F, Hoe J C. Data reorganization in memory using 3D-stacked DRAM. In: Proceedings of ACM/IEEE International Symposium on Computer Architecture. 2015, 131–143

[27] Jevdjic D, Volos S, Falsafi B. Die-stacked DRAM caches for servers: hit ratio, latency, or bandwidth? have it all with footprint cache. In: Proceedings of ACM/IEEE International Symposium on Computer Architecture. 2013, 404–415

[28] Oskin M, Loh G H. A software-managed approach to die-stacked DRAM. In: Proceedings of International Conference on Parallel Architecture and Compilation. 2015, 188–200

[29] Mekkat V, Holey A, Yew P C, Zhai A. Managing shared last-level cache in a heterogeneous multicore processor. In: Proceedings of International Conference on Parallel Architectures & Compilation Techniques. 2013, 225–234

[30] Lee M, Kim S. Performance-controllable shared cache architecture for multi-core soft real-time systems. In: Proceedings of IEEE International Conference on Computer Design. 2013, 519–522

[31] Pan A, Pai V S. Runtime-driven shared last-level cache management for task-parallel programs. In: Proceedings of International Conference for High Performance Computing, Networking, Storage and Analysis. 2015, 1–12

[32] Albericio J, Ibáñez P, Viñals V, Llabería J M. The reuse cache: downsizing the shared last-level cache. In: Proceedings of IEEE/ACM International Symposium on Microarchitecturee. 2013, 310–321

[33] Loh G H, Hill M D. Supporting very large DRAM caches with compound-access scheduling and MissMap. IEEE Micro, 2012, 32(3): 70–78

[34] Loh G H, Hill M D. Efficiently enabling conventional block sizes for very large die-stacked DRAM caches. In: Proceedings of IEEE/ACM International Symposium on Microarchitecture. 2011, 454–464

[35] Dong H W, Seong N H, Lee H H S. Pragmatic integration of an SRAM row cache in heterogeneous 3-D DRAM architecture using TSV. IEEE Transactions on Very Large Scale Integration Systems, 2013, 21(1): 1–13

[36] Chen Q, Zheng L, Guo M. Adaptive demand-aware work-stealing in multi-programmed multi-core architectures. Concurrency & Computation: Practice & Experience, 2016, 28(2): 455–471

[37] Suo G, Yang X. System level speedup oriented cache partitioning for multi-programmed systems. In: Proceedings of IFIP International Conference on Network and Parallel Computing. 2009, 204–210

[38] Kirovski D, Lee C, Potkonjak M, Mangione-Smith W H. Applicationdriven synthesis of memory-intensive systems-on-chip. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 1999, 18(9): 1316–1326

[39] Sim J, Loh G H, Sridharan V, O'Connor M. A configurable and strong RAS solution for die-stacked DRAMcaches. IEEEMicro, 2014, 34(3): 80–90

[40] Lin B, Li S, Liao X, Zhang J. Leach: an automatic learning cache for inline primary deduplication system. Frontiers of Computer Science, 2014, 8(2): 175–183.

[41] Chou C, Jaleel A, Qureshi M K. CAMEO: a two-level memory organization with capacity of main memory and flexibility of hardwaremanaged cache. In: Proceedings of IEEE/ACM International Symposium on Microarchitecture. 2014, 1–12

[42] Ou J, Patton M, Moore M D, Xu Y, Jiang S. A penalty aware memory allocation scheme for key-value cache. In: Proceedings of International Conference on Parallel Processing. 2015, 530–539

[43] Woo H D, Seong N H, Lewis D L, Lee H H S. An optimized 3D-stacked memory architecture by exploiting excessive, high-density TSV bandwidth. In: Proceedings of the 16th IEEE International Symposium on High Performance Computer Architecture. 2010, 1–12

[44] Loh G H. Extending the effectiveness of 3D-stacked DRAM caches with an adaptive multi-queue policy. In: Proceedings of IEEE/ACM International Symposium on Microarchitecture. 2009, 201–212

[45] Jiang L, Liu Y, Duan L, Xie Y, Xu Q. Modeling TSV open defects in 3D-stacked DRAM. In: Proceedings of IEEE International Test Conference. 2010, 174–182

[46] Hennessy J L, Patterson D A. Computer Architecture: A Quantitative Approach. Elsevier, 2011

[47] Li S, Cheng B, Gao X, Qiao L, Tang Z. Performance characterization of SPEC CPU2006 benchmarks on Intel and AMD platform. In: Proceedings of IEEE International Workshop on Education Technology & Computer Science. 2009, 116–121

[48] Sim J, Loh G H, Kim H, O'Connor M. A mostly-clean DRAM cache for effective hit speculation and self-balancing dispatch. In: Proceedings of IEEE/ACM International Symposium on Microarchitecture. 2012, 247–257

[49] Begum R, Hempstead M. Power-agility metrics: measuring dynamic characteristics of energy proportionality. In: Proceedings of IEEE International Conference on Computer Design. 2015, 643–650

RIGHTS & PERMISSIONS

2018 Higher Education Press and Springer-Verlag GmbH Germany, part of Springer Nature