Finding and Fixing Performance Pathologies in Persistent Memory Software Stacks

Xu, Jian; Kim, Juno; Memaripour, Amirsaman; Swanson, Steven

doi:10.1145/3297858.3304077

Cited by 42 publications

(19 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We report the absolute performance for the baseline file system in each category and relative throughput for SplitFS. Despite our best efforts, we were not able to run Strata on these large applications; other researchers have also reported problems in evaluating Strata [32]. We evaluated Strata with a smaller-scale YCSB workload using a 20GB private log.…”

Section: Performance On Data-intensive Workloadsmentioning

confidence: 99%

“…Unlike prior work like Aerie, which used the kernel only for coarsegrained operations, or Strata, where all operations are in user-space, SplitFS routes all metadata operations to the kernel. While FLEX [32] invokes the kernel at a fine granularity like SplitFS, it does not provide strong semantics such as synchronous, atomic operations to applications. At a high level, the SplitFS architecture is based on the belief that if we can accelerate common-case data operations, it is worth paying a cost on the comparatively rarer metadata operations.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

SplitFS

Kadekodi

Lee

Kashyap

et al. 2019

Proceedings of the 27th ACM Symposium on Operating Systems Principles

122

View full text Add to dashboard Cite

We present SplitFS, a file system for persistent memory (PM) that reduces software overhead significantly compared to state-of-the-art PM file systems. SplitFS presents a novel split of responsibilities between a user-space library file system and an existing kernel PM file system. The user-space library file system handles data operations by intercepting POSIX calls, memory-mapping the underlying file, and serving the read and overwrites using processor loads and stores. Metadata operations are handled by the kernel PM file system (ext4 DAX). SplitFS introduces a new primitive termed relink to efficiently support file appends and atomic data operations. SplitFS provides three consistency modes, which different applications can choose from, without interfering with each other. SplitFS reduces software overhead by up-to 4× compared to the NOVA PM file system, and 17× compared to ext4 DAX. On a number of micro-benchmarks and applications such as the LevelDB key-value store running the YCSB benchmark, SplitFS increases application performance by up to 2× compared to ext4 DAX and NOVA while providing similar consistency guarantees. CCS Concepts• Information systems → Storage class memory; • Hardware → Non-volatile memory; • Software and its engineering → File systems management;

show abstract

Section: Performance On Data-intensive Workloadsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

SplitFS

Kadekodi

Lee

Kashyap

et al. 2019

Proceedings of the 27th ACM Symposium on Operating Systems Principles

122

View full text Add to dashboard Cite

show abstract

“…Thus, the upcoming construction of larger MCC infrastructures is expected to be equipped with an array of persistent memory devices co-located with DRAM on each node or shared among all the nodes via highspeed interconnects such as Gen-Z [6] and Infiniband to improvise MCC [6]- [8]. In MCC, the nodes are equipped with non-volatile memories (NVMs), such as Intel Optane DC Persistent Memory (PM) which offers high capacity at low cost, byte-addressability, low idle power, persistence, and performance closer to DRAM than SSD or disks [10]- [12].…”

Section: Introductionmentioning

confidence: 99%

MOSIQS: Persistent Memory Object Storage With Metadata Indexing and Querying for Scientific Computing

et al. 2021

View full text Add to dashboard Cite

Scientific applications often require high-bandwidth shared storage to perform joint simulations and collaborative data analytics. Shared memory pools provide a chance to satisfy such needs. Recently, a high-speed network such as Gen-Z utilizing persistent memory (PM) offers an opportunity to create a shared memory pool connected to compute nodes. However, there are several challenges to use scientific applications on the shared memory pool directly such as scalability, failure-atomicity, and lack of scientific metadata-based search and query. In this paper, we propose MOSIQS, a persistent memory object storage framework with metadata indexing and querying for scientific computing. We design MOSIQS based on the key idea that memory objects on PM pool can live beyond the application lifetime and can become the sharing currency for applications and scientists. MOSIQS provides an aggregate memory pool atop an array of persistent memory devices to store and access memory objects to accelerate scientific computing. MOSIQS uses a lightweight persistent memory key-value store to manage the metadata of memory objects, which enables memory object sharing. To facilitate metadata search and query over millions of memory objects resident on memory pool, we introduce Group Split and Merge (GSM), a novel persistent index data structure designed primarily for scientific datasets. GSM splits and merges dynamically to minimize the query search space and maintains low query processing time while overcoming the index storage overhead. MOSIQS is implemented atop of PMDK. We evaluate the proposed approach on many-core server with an array of real PM devices. Experimental results show that MOSIQS gains a 100% write performance improvement and executes multi-attribute queries efficiently with 2.7× less index storage overhead offering significant potential to speed up scientific computing applications.

show abstract

“…Their approaches commonly include bypassing the traditional block layer and the page cache layer to reduce unnecessary software overheads in NVM file systems. Furthermore, researchers have proposed user-level file systems based on NVM to remove the whole IO stack overhead within the OS kernel [6,15,18,31,34]. Overcoming the limitations of existing file systems, which are mainly designed for slow block devices, these NVM-aware file systems have achieved notable performance improvements.…”

Section: Introductionmentioning

confidence: 99%

Libpubl

Jung

Choi

Han

2021

Proceedings of the 13th ACM Workshop on Hot Topics in Storage and File Systems

View full text Add to dashboard Cite

With recent advancement in NVM technologies, non-volatile main memory (NVMM) has been highlighted as promising systems for large scale data centers. By utilizing memorymapped IO, researchers have attempted to provide fast IOs without kernel mode switches and complex kernel IO layers. Even though they improve the performance by taking advantage of NVMM characteristics, memory-mapped IOs still require additional user-level logging techniques to guarantee write atomicity and file system integrity. We propose a user-level library file system, Libpubl, which is designed to minimize the write amplification overhead of traditional logging by exploiting persistent user buffers as logs. Libpubl ensures atomic updates of data and improves the performance and the scalability. Compared to the state-of-the-art NVM file systems, Libpubl increases the performance by 50∼120% in Fio benchmark. CCS CONCEPTS• Software and its engineering → File systems management; • Information systems → Storage class memory;• Hardware → Memory and dense storage.

show abstract

Finding and Fixing Performance Pathologies in Persistent Memory Software Stacks

Cited by 42 publications

References 35 publications

SplitFS

SplitFS

MOSIQS: Persistent Memory Object Storage With Metadata Indexing and Querying for Scientific Computing

Libpubl

Contact Info

Product

Resources

About