• Corpus ID: 76667960

Basic Performance Measurements of the Intel Optane DC Persistent Memory Module

@article{Izraelevitz2019BasicPM,
  title={Basic Performance Measurements of the Intel Optane DC Persistent Memory Module},
  author={Joseph Izraelevitz and Jian Yang and Lu Zhang and Juno Kim and Xiao Liu and Amirsaman Memaripour and Yun Joon Soh and Zixuan Wang and Yi Xu and Subramanya R. Dulloor and Jishen Zhao and Steven Swanson},
  journal={ArXiv},
  year={2019},
  volume={abs/1903.05714}
}
Scalable nonvolatile memory DIMMs will finally be commercially available with the release of the Intel Optane DC Persistent Memory Module (or just "Optane DC PMM. [] Key Result When used in this Memory (or cached) mode, Optane DC memory has little impact on applications with small memory footprints. Applications with larger memory footprints may experience some slow-down relative to DRAM, but are now able to keep much more data in memory. When used under a file system, Optane DC PMMs can result in…
Understanding and Improving Persistent Transactions on Optane (TM) DC Memory
TLDR
A large throughput difference is found, which emphasizes the importance of choosing the best durability domain for each application and system, and confirms that recently published persistent transactional memory algorithms are able to scale, and that recent optimizations for these algorithms lead to strong performance.
Trend of Intel Nonvolatile Memory Technology
TLDR
To take advantage of the performance characteristics of this device, operating systems and applications require new approaches, and this change in computing environments will require a significant number of researches in the future.
System evaluation of the Intel optane byte-addressable NVM
TLDR
This work provides an in-depth evaluation of the first commercially available byte-addressable NVM -- the Intel Optane® DC™ persistent memory and employs two fine-grained allocation policies to control traffic distribution between DRAM and NVM.
Lessons learned from the early performance evaluation of Intel optane DC persistent memory in DBMS
TLDR
This paper provides extensive performance evaluations on a recently released NVM device, Intel Optane DC Persistent Memory (PMem), under different configurations with several micro-benchmark tools and shares some recommendations for future DBMS design with PMem.
Performance Evaluation of the Intel Optane DC Memory With Scientific Benchmarks
TLDR
The performance of scientific application from domains of quantum chemistry and computational astrophysics is studied and it is seen that PMM is superior to IMDT in almost all benchmarks, to no surprise.
Large-scale in-memory analytics on Intel® Optane™ DC persistent memory
TLDR
One of the first experimental studies on characterizing Intel® Optane™ DC PMM's performance behavior in the context of analytical database workloads is presented, revealing interesting performance tradeoffs that can help guide the design of next-generation OLAP systems in presence of persistent memory in the storage hierarchy.
Characterizing the performance of intel optane persistent memory: a close look at its on-DIMM buffering
TLDR
Interestingly, in addition to write coalescing, the write buffer delivers lower than read and consistent write latency regardless of the working set size, the type of write, the access pattern, or the persistency model.
Unexpected Performance of Intel® Optane™ DC Persistent Memory
TLDR
It is found that Intel's persistent memory is highly sensitive to data locality, size, and access patterns, which becomes clearer by optimizing both virtual memory page size and data layout for locality.
Understanding and Improving Persistent Transactions on Optane™ DC Memory
TLDR
A large throughput difference is found, which emphasizes the importance of choosing the best durability domain for each application and system, and confirms that recently published persistent transactional memory algorithms are able to scale, and that recent optimizations for these algorithms lead to strong performance.
A Case Against Hardware Managed DRAM Caches for NVRAM Based Systems
TLDR
This work analyzes the performance of such DRAM caches on real hardware using a broad range of synthetic and real-world benchmarks and shows that software based techniques are necessary for orchestrating the data movement between DRAM and PMM for such workloads to take full advantage of these new heterogeneous memory systems.
...
...

References

SHOWING 1-10 OF 22 REFERENCES
HME: A lightweight emulator for hybrid memory
TLDR
Experimental results demonstrate that the average emulation errors of NVM read and write latencies are less than 5% in HME, which is much lower than Quartz, and the application performance overhead in Hme is one order of magnitude higher than Quartz.
Mnemosyne: lightweight persistent memory
TLDR
In tests emulating the performance characteristics of forthcoming SCMs, Mnemosyne can persist data as fast as 3 microseconds and can be up to 1400% faster than alternative persistence strategies, such as Berkeley DB or Boost serialization, that are designed for disks.
NOVA-Fortis: A Fault-Tolerant Non-Volatile Main Memory File System
TLDR
An NVMM-optimized file system called NOVA-Fortis is described that is both fast and resilient in the face of corruption due to media errors and software bugs and outperforms reliable, block-based file systems running on NVMM by 3x on average.
Atlas: leveraging locks for non-volatile memory consistency
TLDR
This paper identifies failure-atomic sections of code based on existing critical sections and describes a log-based implementation that can be used to recover a consistent state after a failure, and confirms the ability to rapidly flush CPU caches as a core implementation bottleneck and suggest partial solutions.
Aerie: flexible file-system interfaces to storage-class memory
TLDR
Aerie is presented, a flexible file-system architecture that exposes storage-class memory to user-mode programs so they can access files without kernel interaction and enables applications to optimize the file system interface.
NV-Tree: Reducing Consistency Cost for NVM-based Single Level Systems
TLDR
NV-Tree, a consistent and cache-optimized B+Tree variant with reduced CPU cacheline flush, and NV-Store, a key-value store based on NV- tree, are implemented and evaluated on an NVDIMM server.
SCMFS: A file system for Storage Class Memory
  • XiaoJian WuSheng QiuA. Reddy
  • Computer Science
    2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC)
  • 2011
TLDR
This paper proposes a new file system, called SCMFS, which is implemented on the virtual address space, which utilizes the existing memory management module in the operating system to do the block management and keep the space always contiguous for each file.
A Workload Characterization of the SPEC CPU2017 Benchmark Suite
  • Ankur LimayeTosiron Adegbija
  • Computer Science
    2018 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)
  • 2018
TLDR
This paper extensively characterize the SPEC CPU2017 applications with respect to several metrics, such as instruction mix, execution performance, branch and cache behaviors, and presents detailed analysis to enable researchers to intelligently choose a diverse subset of the CPU2017 suite that accurately represents the whole suite, in order to reduce simulation time.
AndroStep: Android Storage Performance Analysis Tool
TLDR
The Android storage performance analysis Tool (AndroStep) is introduced which is specifically designed for characterizing and analyzing the behavior of the IO subsystem in Android based devices.
SPEC CPU2006 benchmark descriptions
On August 24, 2006, the Standard Performance Evaluation Corporation (SPEC) announced CPU2006 [2], which replaces CPU2000. The SPEC CPU benchmarks are widely used in both industry and academia [3].
...
...